a sociophonetic study of selected morphophonetic ...€¦ · egyptian arabic, from about the mid of...

الجزء الثاني 7102العدد الثامن عشر لسنة لعلمي في اآلداب مجلة البحث ا

1

A Sociophonetic Study of Selected Morphophonetic

Alternations in the Egyptian Folk Epic Beibers

Presented by

Esraa Bahaa El-Din Mohamed Rashad

Department of English Language and Literature

Faculty of Women- Ain Shams University

Supervised by

Dr. Heba Mohamed Ibrahim Aboul-Enein Dr. Heba Ahmed Zaytoun

Assistant Professor of Linguistics

Department of English Language and

Literature

Faculty of Women-Ain Shams University

T Lecturer of Linguistics Department of English Language and

Literature

Faculty of Women-Ain Shams University


2

Abstract

This paper examines selected phonetic and morphological phenomena in early

Egyptian Arabic, from about the mid of the 13th century up till, approximately, the

end of the 16th

century. It attempts to reveal how certain Egyptian Arabic words

deviated from Classical Arabic (CA) pronunciation. The data are primarily extracted

from the epic of Al-Zāhir Beibers, which flourished during that era. In order to reach

this end, the paper adopts a synchronic-diachronic perspective to data analysis under

the theory of Lexical Phonology, within the broader framework of Sociophonetics.

The selected sound changes include chosen cases of glottal stop alternations, vowel

shortening, emphatic spread, and conversion of interdentals. The findings show that

some of the depicted alternations were used as variants in Al-Jazeera Arabic dialects.

Others are presumed to be a result of false analogical application of originally

Classical Arabic rules, the Coptic influence on Egyptian Arabic, and for ease of

articulation. Moreover, the results reflect the adequacy of Lexical Phonology as an

analytical tool when dealing with such morphophonological alternations.

Key words:

16th

century Egyptian Arabic- Epic of Beibers- Sociophonetics- Lexical Phonology -

phonemes alternations

1- Inventory of consonants used throughout the paper (i.e. CA&EA)

Place

Bilabial

Labio-

dental

Inter-

dental

Denta

l

Dento-

Alveolar

Palatal

post-

Alveola

r

Vela

r

Uvular

Pharyn-

geal

Laryng

-eal/ Glottal


3

Manner

Plosive

b ب

t ت d د

k ك g ج

(EA)

q ق

ʔ ء

Plosive

Emphati

c

T ط D ض

Affricat

e

ʤ, j

(CA)

dialectal

EA ) ج

Fricativ

e

f ف

th ث

TH ر z ز

s س

sh ش

gh غ X خ

ʕ ع H ح

h ه

Fricativ

e

Emphat

ic

TH ظ

Z

ز مفخمة(EA

Only) S ص

Trill r ر

Lateral l

ل

Nasal m م

n ن

Glide

y ي

w

و

(Adapted from Newman, 2006, p.2)

Further Notes on Consonants - All the phonemic symbols above are used also in transliteration, except for [‘] for

/ʔ/ ء, [‗] for /ʕ/ ع, and [kh] for /x/ خ

-Underlining is used to denote emphasis spread.

2- Symbols of Vowels

Front Back

High

(closed)

/i/, /ii/ /u/, /uu/


4

Mid /ee/ /oo/

Low

(open)

/a/-/a/, /aa/-/aa/ /ā/, /āā/- /ā/, /āā/,

(Adapted from Watson, 2002, pp.21- 23)

3- List of Abbreviations and Symbols

Abbreviation full form Symbol Meaning

Antepenult. Antepenultimate ] word Final

position

Penult. Penultimate [ word/phrase

initial position

C, V consonant, vowel / If / Under the

condition

(F)VS (final) vowel

shortening

, < alternates to /

becomes

HVD high vowel

deletion rule

[ ] morphological

unit/ allophone

O, C, N onset, coda, nucleus ( ) Optionality

VL vowel lengthening µ Mora

SL source language σ syllable level

CA Classical Arabic ` Stress

SA Standard Arabic * ill-formedness

EA Egyptian Arabic

UR underlying Form

PF phonetic Form

LP Lexical Phonology

4- Arabic Morphological Categories: Abbreviations and Arabic translations

Category Arabic Translation

Imperfective (Imperf.) Perfective

(perf.)

اىبضاىفؼو -اىفؼو اىضبسع

Masculine (masc.), Feminine (fem.) اىؤث -اىزمش

Singular (sing.), Plural (plu.) اىدغ -اىفشد

Active Participle (AP) اع اىفبػو

Preposition (prep( حشف اىدش

Hollow verb اىفؼو االخف

Imperative verb (Imp.) اىفؼو األش

Adjective (Adj.) طفخ


5

0. Introduction

Egypt manifests a distinguishably rich model of language variation, which can

be attributed to the ongoing state of change in its social and political life throughout

its long history. Egyptian Arabic has always been a source of attraction for both

Arab and Western linguists, being the most popular dialect in the Arab world.

Nevertheless, studies have been devoted, basically, to contemporary Egyptian

dialect(s) rather than to the early variety. This, consequently, led to a lack of

knowledge concerning Egyptian Arabic path of variation and change. Moreover, in

order to trace the historical phases of a language or a dialect, linguists usually resort

to written records. However, the majority of early Egyptian records reflected

Classical Arabic, rather than representing the spoken variety. Nonetheless, fewer

Egyptian documents, written in a language that mixes CA with the colloquial variety,

survived. Al-Zāher Beibers (1996) is considered to be one of those scarce sources or

representatives of Egyptian Arabic, which was spoken at the Mamluk and Ottoman

Egypt.

0.1. Sirat Al-Zāher Beibers

Sirat Al-Zāher Beibers is a folk epic tale that narrates the story of Sultan

Beibers Al-Bunduqdāāriy. Beibers was the 4th baHariy اىجحش (maritime) Mamluk

ruler of Egypt and the Levant from 1260 till his death in 1277 (Shalaby, 1992, pp.

32-33). According to Al- GhiTāāni (1996), Beibers‘ wisdom, justice and rationality

Noun (n.) اع


6

made him one of the most loved and respected rulers throughout the history of

Egypt. In addition, Egyptians‘ great appreciation for Sultan Beibers led them to

weave folk stories about him. Thus, the epic aimed at describing his own and his

companions‘ adventures, championships, relationships, and different circumstances

of his sultanate, in which reality was mixed with imagination. The epic was recited

for ages in popular Egyptian cafes that were named after it as Al-Zāheriyya before its

completion during the Ottoman era. He added that the epic provides a linguistic

evidence of the colloquial dialect spoken during the Mamluk era (p.6). Likewise,

‗ĀTiyyāh (2001) assers that one of the unique elements of Beibers is its successful

contribution to preserving the Egyptian dialect and expressions used during that time

(n. p).

0.2. Language in Egypt Ever Since the Dawn of History

It is believed that the history of language in Egypt extends thousands of years

back in time. According to Allen (2004) ―Egyptian first appeared in writing shortly

before 3000 BC and remained in active use until the eleventh century AD‖ (p. i).

Scholars have divided Egyptian language phases chronologically, based on the types

of scripts they discovered. Thus, the first type was Hieroglyphs; the second and third

types were Hieratic, and Demotic. Coptic came as the fourth and the last descendant

of ancient Egyptian language. It was the indigenous language of Egyptians up till the

advent of Islamic state in the 7th

century, with its official Arabic language. According

to Haikal (2003), ―Arabic must have been initially acquired next to Coptic‖. However,

by the ―twelfth century Arabic had become the main written language in the church,

indicating that the population could no longer understand Coptic‖ (pp. vii-ix). Hence,

Coptic, as well as its parent ancient Egyptian were assumed to have left traces on

Egyptian Arabic. Moreover, the Arabic spoken by Egyptians has possibly been

affected by Al-Jazeera Arabic dialects, amongst other languages. Hence, the paper

studies some of these aspects under the rubric of sociolinguistics.

1.0. Sociolinguistics

Sociolinguistics is basically concerned with investigating language within its

social context. Wodak, Johnstone, and Kerswill (2011) heralded that sociolinguistics

has ―endorsed the view that language could not be adequately understood without

taking many layers of social context into account, be it the situational context of

utterances, the geographical origin of speakers, the age, the gender, social class, ...‖

(pp. 1-2). Thus, language variation is a consequence of an intersection between

linguistic and extra-linguistic factors.


7

1.1. Arabic Situation

Variation shows to be a central characteristic of Arabic speaking countries. This

consequently, as Haeri (2000) confirmed, has led to the emergence of the more

specialized ―Arabic Sociolinguistics‖ in the mid of the 1970s (p. 62). Arabic

sociolinguistics sought to evaluate and analyze speaker linguistic choices in line with

the situational and communal contexts. This was enriched by utilizing sociolinguistic

notions and ―trends in reseach on diaglossia, code-switching, gendered discourse,

language variation and change, and language policies in relation to Arabic‖

(Bassiouney, 2009, p. 6).

Examining linguistic variation in relation to the Arabic community/speakers

started as early as medieval ages. Miller and Caubet (2010) argued that Medieval

Arab grammarians were ―aware of the heterogeneity of the Arabic language‖ (p.

239). This was further affirmed by Bassiouney (2009) who signaled out that the

―interest in the differences in ways people speak is very old‖, referring to Al-Khaliil

Ibn Ahmed Al-Farahiidiy as the first one who mentioned the concept of phonological

variation in his dictionary of kitaabu-l-‘Ayn (p. 5). However, the chief objective was,

then, to preserve Classical Arabic, and refine it rather than to study variation per se,

as a valid characteristic of language (Miller & Caubet, 2010, p. 421).

1.2. Language Change and Sociolinguistics

The state of language change has proved to be in a direct relationship with

time. Guy (1996) asserts that language is essentially ―a dynamic social system, which

is continuously moving, changing, interacting, and working‖ (p. ix). In addition,

Attinasi (1974) elucidated that there are three aspects from which language change

could be approached, namely, ―the origin, the spread, and the regularization of

language differences through time‖ (p. 282). According to Coulmas (2013), ―the fact

of language change forces us to look at instability, deviation and loss of

comprehension across generations and dialects‖ (p.4). This means that language

would not be able to resist deviation from what might at certain times be regarded as

a standard form.

Aitchison (2001) affirms that ―a change tends to sneak quietly into a language,

like a seed, which enters the soil and germinates unseen‖ (p. 66). This was

corroborated by Hickey (2001) explaining that ―[t]he way languages change offers

an insight into the nature of language itself, its internal organisation, and how it is


8

acquired and used‖. He asserts the inevitable relation between language, dynamicity

and ultimately change, sustaining that ―given the dynamic nature of language,

change is ever present‖ (p. 1). Henceforth, dynamicity here could be equivalent to

variability, where language variation acts as a preliminary stage for language change.

This, therefore, renders language variation as the substantial key for understanding

language change. Further, Baker (2010) emphasizes that ―exploring the historical

and social contexts of time periods being examined‖ results in a ―more reliable

description and explanation‖ (p.80).

This could be witnessed in English language, for instance, which is well-

recognized for manifesting correlation between linguistic changes and external

factors represented by sociohistorical events. Meyer (2009) maintains that ―various

historical and linguistic events led to changes in English over time‖ (p. 20). This

could be witnessed in the borrowings of either lexical items or linguistic habits that

take place across English history. Algeo (2010) explicates that one of the proposed

triggers of sound change, (for instance), is language contact, which usually happens

when one group of speakers learn the language of another. However, within the

course of attempting to learn the new foreign language, sometimes the native

speakers tend to transmit ―native habits of pronunciation into the newly acquired

language‖ (p. 30). According to Hickey (20 English for example ―has lost

consonantal length, has acquired phonemic voiced fricatives, and has developed

contrastive word stress under Romance influence (p. 5). Some changes, however,

have been attributed to internal factors, others are still mysterious.

1.3. Language and Society in Mamluk and Ottoman Egypt

Lane-Poole (1901) mentions that during this era, ―the population of Egypt was

sharply divided into two classes who had little in common with each other‖. One

division was ―the military oligarchy‖, and the other represents ―the mass of the

Egyptians‖ (pp. 252-253). Al-Māqriiziy (2007) describes the diverse layers of the

Egyptian mass in detail, dividing the society into seven classes (pp.147-149), naming

the first group as ― اىذىخأو ‖ ‘ahlu-l- dawlah (the ruling class). The second and the

third are: ―او اىغبس اىزدبس أى اىؼخ ر اىشفبخ‖, ‘ahlu-l-yasaar min-al-tuʤāār

wa-‘uli-l-niʕmah min THawi-l-rāfaahiyah (Well-off merchants), and ― اىجبػخ

اىزدبسزعط اىحبه ‖ ‘al-baa‗ah wa-hum mutawāsiTu-l-Haal min-al-tuʤāār (vendors

who are financially moderate). The fourth and the fifth are: ―أو اىفيح‖ ‘ahlu-l-filH

(Peasants and villagers), and ―خو اىفقشاء طالة اىؼي‖ ʤullu-l-fuqārāā‘ wa-Tullāābu-l-

‗ilm (Poor people and students). The sixth and the seventh groups consist of ― أسثبة"

-arbaabu-l-Sānāā‘i‗ wa-l-‘uʤarāā‘ wa-‘aSHāābu-l‘ اىظبئغ األخشاء أطحبة اى

mihan (craftsmen and wage earners), and the less priviligded class of ر اىحبخخ


9

-THawu-l-Haaʤati wa-l-maskanah (panhandlers and poor fellows) (pp.147 ‖اىغنخ"

149). Hence, such diversity of the Egyptian society is reflected on the language

spoken by its members. According to Selim (1957), the foreign ruling class, for

example, are of Turkic origins, and spoke Turkic languages (i.e. Turkish and Jerkisi);

whereas Egyptians speak colloquial Arabic. He further assures the co-existence of

colloquial Arabic, which can collectively be referred to as Egyptian Arabic, and

foreign languages, mainly used by the ruling class. This is in addition to Classical

Arabic, which is restricted to formal religious, scientific, literary, and poetic

writings, etc. (p. 21).

1.4. A Brief Description of Arabic Morphological System and Syllable Structure

Fullwood and O‘Donnell (2013) state that ―the predominant mode of Semitic

word formation is non-concatenative‖ (p. 21). Arabic, being a descendant of Semitic

family, conforms to this mode. Ryding (2005) illustrates that Arabic morphological

system is marked by a ―discontinuous morphemes‖ pattern. She further explaines the

way Arabic morphology is constituted, where the ―system of consonant roots …

interlock[s] with patterns of vowels (and sometimes certain other consonants) to form

words, or word stems‖ (p.45).

Crystal (2008) defines ―root‖ as ―a sequence of consonants‖ (p. 418), i.e. in

Arabic [F.ʔ.L]. Words after intervening vowels could be measured against templatic

scales. According to Haywood & Nahmad (1965), the definite number of derived

forms is ―open to dispute‖ (p. 151). Nevertheless, the first 10 forms of Arabic verbs

are, generally, widely discussed in the literature because of their frequent use. The

most frequent, however, is the basic form I [Faʕa(i)(u)L], followed by the form II

[faʕʕala], III [faaʕala], IV [ʔafʕala], V [tafaʕʕala], VI [tafaaʔala], VII [ʔinfaʕala], VIII

[ʔiftaʕala], IX [fʕalla], and X [ʔistafʕala]. Thence, what primary turns roots to words

is the intervening vowels for both nouns and verbs, which is the case in CA, as well

as its dialects including Egyptian Arabic. Moreover, Abdil-Tawwaab (2000, p. 55) &

Ryding (2005, p. 36) amongst others, have listed five syllable structures for CA.

Mitchell (1956, p. 110) confirmed the presence of the same five syllables for EA as

follows: CV, CVV, CVC, CVVC, and CVCC.

1.4. Previous Studies on EA

This section briefs on some of the previous related studies of EA. As for

historical oriented studies, the researcher has found few studies conducted on early

phases of Egyptian Arabic. Allahʤah Al-‘Aamiyyah Al-MāSriyyāh fi-l-Qārn Al-


11

Haadi 'Ashar Al-Hiʤriy by ‗Abdi-l-Tawwaab (n.d.), and Egyptian Arabic in the

seventeenth century: A study and edition of Yūsuf Al-Maghribiy’s Dafʿ Al-’ISr ʿan

Kalaam Ahl MiSr by Zack (2009) are two of the most influential studies that

investigated the linguistic characteristics of early EA. Both studies were carried out

on Daf’ Al-’iSr, a manuscript that is regarded as one of the most precious linguistic

documents that studied the Egyptian dialect of the 16th

& 17th centuries. Another

distinguished study is Al -Lahʤatu-l-MiSriyyatu-l- FāāTimiyyah (1999, 2014) by

‗ĀTiyya Sulaymaan Ahmed. The study is classified as a historical linguistic study

that, basically, concentrated on describing the Egyptian dialect during the Fatimid

era. In addition, a recent study, entitled A Sociophonetic Study of Selected words in

Al-Qawl Al-Muqtadab Fiima Waafaqa Lughat Ahl Misr, studied Al-Qawl Al-

Muqtadab, an abridged version of Daf’-l-’iSr, as representative of Egyptian Arabic in

the Early Ottoman Period by Aboul Enein (2013). The research examined the

morphophonological features of EA words during that era utilizing the theory of

Lexical Phonology.

Woidich and Behnstedt‘s Atlas (1985) is one the most significant dialectological

studies that led to the advance of Arabic sociolinguistics in current time. The study is

known for delving deep into investigating Egyptian dialects based on geographical

distribution. The Atlas has been described by Eissele (1987) as ―the best

dialectological work to have been carried out on Arabic dialect‖ (p. 221), as it drew

an accurate map of various unstudied regions of Egypt (i.e. Delta and the Nile valley,

as well as the oases). Further researches have been concerned with providing a

colloquial Egyptian lexicon, such as Badawi and Hinds‘ A Dictionary of Egyptian

Arabic (1973), which is considered one of the most comprehensive compilers of EA

lexical items. In addition, Ezzat conducted a number of valuable studies on Egyptian

Arabic from multiple dimensions. Among these studies is: Intelligibility among

Arabic Dilects in 1974, where he examined aspects of EA. Moreover, he investigated

the wider scope of the influence of EA on 2nd

language acquisition (English) in his

Some Phonological Problems involved in Teaching English to Speakers of Egyptian

Arabic (Cairene Dialect) in 1977. Finally, Harrell‘s The Phonology of Colloquial

Egyptian Arabic (1957), and Olmsted and Gamal El-din‘s Cairene Egyptian

Colloquial Arabic (1981) are also significant studies that examined phonology and

morphology of EA.

2. Methodology

2.1. Purpose of the Study

The current study is an attempt to explore the way Egyptians talked at a certain

period of Egypt‘s history. This was carried out by investigating the colloquial variety

as reflected in the folk epic of Al-Zāher Beibers (1996), as a representative of EA


11

spoken at the time of the Mamluk and Ottoman Egypt. Thus, the first radical aim is

to conduct a morphophonological analysis on the EA words selected from Beibers

with respect to the historical social dimension. The second end is to specify the

appropriateness of the theory of LP for the analysis. The third purpose is to specify

the similarities/differences between early EA and CA as well as contemporary EA

and early EA, with respect to the application of certain morphophonological rules in

selected words. This is in addition to tracing the persistence of a given phenomenon

productivity through time.

2.2. Research Questions

The study attempts to answer the following questions:

1- What are the phonological/morphological processes involved in altering Classical

Arabic words into colloquial ones in the selected corpus? Do they exhibit a

systematic or idiosyncratic attitude?

2- In view of the scarcity of studies conducted on Arabic language under Lexical

Phonology, to what extent does such an analytical tool applicable on the data under

investigation?

3- Taking into account adopting a synchronic/diachronic approach to the data, did

the same phonological/morphological alternations detected in the data exhibit

persistence through time?

4- Could the epic of Al-Zaher Beibers be considered as a good representative of the

dialect spoken at the Mamluk and the early Ottoman periods?

2.3. Source of Data

Sirat Al-Zāher Beibers (1996) is the main source of data selected for this

study, from which words exhibiting orthographic variations were extracted as

representative of EA phonological and morphological alternations (i.e. colloquial

variety). It consists of fifty parts falling in five volumes, forming a total of 3000

pages. The text misses all kinds of punctuation (i.e. commas, periods, semi colons,

question marks), and is written in CA mixed with the local colloquial language

used in Egypt during these early periods.

2.4. Corpus, method, and Procedures

The analysis was conducted on 140 words from Beibers. Words were

collected manually, due the lack of a computerized version of the corpus. Thus, the


12

first step was skimming through the epic, where some words encountered were

inferred as colloquial, as reflected in their orthography (i.e. deviating from CA).

Next, the chosen words were looked up for in CA dictionaries, henceforth; they

were detected as alternating forms of the original CA. In addition, most of the

words selected were either repeated multiply throughout the epic or else proved

persistence up till the present time, and/or were detected in other historical sources.

These were traced through baheth.net, which is a search engine that encompasses

five of the most reputable CA dictionaries. These are Lisaanu-l-‘Ārāb, Maqāāyiisu-

l- Lughah, Al-SiHāāH fi-l-Lughah, Al-Qāāmuus Al-MuHiiT, and Al-‘Abaabu-l-

Zaakhir.

The words were checked also through Elixir, which is a CA/SA morphology

analytical tool. It helped revealing word root forms as well as the meanings of

words, in relation to their derivations, and their grammatical categories. It was also

consulted for checking the transcription of few CA words. The data selected was,

then, classified into groups, based on the alternation they exhibited. Furthermore,

some of the categorized words reflected a number of alternations, which would be

pinpointed (i.e. through the transcription only and/or by more illustration), yet the

focus would be given chiefly to one phenomenon in every sub/section. Thence, LP

was used to account for morphophonological processes, which affected CA words,

turning them into colloquial. It is noteworthy that Moraic theory was also

frequently utilized, whenever found helpful for the phonological description of

processes and their representation. In addition, extra-linguistic aspects were

inspected in an attempt to trace the source of alternations under study.

2.5. Theoretical Framework The current paper is a synchronic- diachronic study, conducted within the broad

framework of Sociophonetics under which Lexical Phonology (LP) was employed.

They are explained in the following subsections.

2.5.1. Synchrony-Diachrony

The distinction between synchrony and diachrony was first established by de

Saussure (Crystal, 2008, p.142). Meyer (2009) explained diachronic studies as

―examining the historical development of a language, taking into consideration

changes it has undergone over time". As for synchronic studies, they relate to the

convention of ―investigating a language in its present form as it is currently spoken

and written‖ (p. 20). Hence, synchronic-diachronic integration could be described as

being in a perpetual complementary distributional relationship.

2.5.2. Sociophonetics


13

It is a relatively new interdisciplinary field which draws on the integration and

interaction of three main approaches, namely, sociolinguistics, phonetics, and

phonology. Sociophonetics has sprung from the broader sociolinguistics, and it

addresses variations with more focus on describing and accounting for

phonetic/phonological variations. Foulkes, Scobbie, Watt (2010) reveal that

sociophonetics aimed at the ―development of theoretical models in phonetics and

sociolinguistics, spanning speech production and perception, with a clear focus on

the origin and spread of change‖ (p. 704).

The term "socio-phonetic" was first recorded by Deshaies-Lafontaine in his

dissertation on variation in Canadian French (1974). Since the mid of the 1990s,

‗sociophonetics‘ studies have witnessed a considerable bounce as well as expansion

(Foulkes et al., 2010, p. 703). Foulkes and Docherty (2006) describe the discipline‘s

borderlines as ―become(ing) increasingly porous‖. They explain that ―sociophonetic

research now amalgamates theories and methods not only from phonetics and

sociolinguistics but also from related fields including … theoretical linguistics,

phonology, and computational linguistics‖ (p.704).

According to Foulkes et al. (2010), sociophonology was used synonymously

with sociophonetics by Dressler and Wodak (1982) in their variationist study of

Viennese German in 1982 (p. 703). Ohala (1990) further clarified the integrational

relationship between phonetics that is the manner in which speech sounds are

produced and perceived ―doing‖, and phonology that is ―how and why‖ speech

sounds are organized and structured in a certain way ―knowing‖ (p. 155).

2.5.3. Lexical Phonology The theory of Lexical phonology is one of the descendants of the school of

Generative grammar (Carr, 1993, p.103). LP was launched by Mohanan (1982,

1986) and Kiparsky (1982), focusing on the relationship between phonology and

morphology. According to Mohanan (1986), "phonology and morphology… are

inputs to each other…", where "the output of phonological operations may undergo

morphological operations, and may then undergo further phonological operations"

(p.8).

2.5.3.1. Lexical vs. post lexical modules

The core assumption of the theory of Lexical Phonology (LP) lies in the

distinction between lexical and postlexical modules. Typically speaking, the

division indicates that rules, which apply word internally, will be assumed to be

supplied by morphological information, hence, occurring within the lexicon.


14

Mohanan (1986) explains that ―A rule application requiring morphological

information must take place in the lexicon‖ (p.9), e.g., /g/ sound deletion in the

word‘sign’ before a nasal, where /g/ /ᴓ/ / ______ [+nasal]] (p. 22). In case

the rule operated across words, then this would designate postlexicallity (p.10). One

example of postlexical rules is the flapping of the stop /t/ both within word

boundary e.g. bu/ɺ/er and across word boundary e.g. See you [ɺ]omorro. Hence,

flapping takes place as a result of the presence of the appropriate structural

description, namely, [-cont., +alveolar, -nasal] (i.e. /t/) between [+stress] and [-

stress] (Stabler, 2007, p. 26).

There is a specific order for these rules to take place, where lexical rules are

expected to operate before postlexical ones. That is why "postlexical operations are

blind to the internal structure of words" (p.14). McMahon (2000) confirmed that

―lexical and postlexical rules display distinct clusters of properties, and are subject

to different sets of constraints‖ (p.5). Moreover, McCarthy (1999) spots some of the

basic characteristics of both modules as seen below:

(p. 55)

The lexical-postlexical categorizaation of rules could, thus, be identified via

being measured against the previous criteria. Hence, in case, for example, a given

process exhibited signs of cyclicity, operation in a derived environment, limitation to

word boundary, results in a contrastive feature as well as applies prior to postlexical

rules, then it could be realized as a lexical rule. Another significant sign that helps

decide on the nature of a rule is the Structure presevering (SP) principle, which,

according to Kiparsky (1985), is ―a result of constraints formulated over the entire

lexicon‖ (p. 87). This was further explained by Watson (2002) as: ―lexical processes

are structure preserving‖ as long as ―they don‘t introduce new segment‖ that is not

part of the underlying inventory of the language‖ (p. 201). A case in point could be,

according to Mohanan (1986), the American Flapping, which is not present in English

phonemic inventory.

In addition, Mohanan (1986) elucidates that rules applying ―solely in the

postlexical stratum, such as ―Aspiration, Flapping, Nasalization of vowels before

nasals, Deletion of r in non-rhotic accents‖ are non-exceptionally productive and


15

therefore are assigned to the postlexical stratum. Nevertheless, rules embracing ―the

packet of Shortening-Lengthening-Vowel Shift-Diphthongization, g Deletion, [n]

Deletion‖ were located by him in a lower rank, which postulates their affiliation to the

lexical stratum (p. 56).

2.5.3.2. Levels and Cyclicity

One of the focal ends of LP is defining the nature of the relationship between the

phonological and the morphological components. Mohanan (1986) states that ―a

subset of phonological rule applications takes place in the lexicon, in tandem with the

morphological operations‖ (p.8). Maye (2000) explaines that LP ―is a derivational

theory which posits various strata in which morphemes are concatenated and

phonological rules may apply‖ (p. 78). This, therefore, assures the idea of interaction

of both componenets, which takes place following a certain order, giving rise to

‗Cyclicity‘. According to Mohanan (1986), the term designates ―a single pass through

the whole rule system‖ (p.23). As a result, examining a rule application would

consistently take place through every single step of the derivation within the lexical

module. He adds that cyclicity is generated out of ―a consequence of the interaction

between the lexical strata and the phonological rule system‖ (p.48).

Cyclicity, additionally, encompasses a constraint that is called ―Bracket Erasure

Convention‖ (BEC). The principle is known to work on deleting ―the (internal)

brackets at the end of all affixation processes‖, after the completion of

―morphological and phonological operations at each level of the lexicon‖ (Carr,

p.176, 1993). One example that well-demonstates both cyclicity and BEC is the rule

of Trisyllabic Vowel Shortening (TSS). The steps of derivation in the lexical module

are depicted by Mohanan (1986) as follows: 1) UR: [diviin] [iti], 2) Affixation:

[[diviin] [iti]], TSS: [[divin][iti]] (i.e. cyclicity), 4) Lexical Representation: [divinity]

(p. 10). Thus, cyclicity is reflected in the subsequential phonological effect of vowel

shortening resulting, primarily, from the morphological process of [iti] suffixation.

This corroborates the state of interaction taking place between both components of

phonology and morphology.

2.5.3.3. Lexical Phonology, the Moraic Model and Arabic Syllable Structure

Lexical Phonology (LP) was employed with reference to the Moraic Model.

Such an approach of utilizing lexical phononology and moraic model has been

adopted by Kiparsky (2003). Such an approach has also been implemented by

Watson (2002), whose model was followed through the data analysis. According to

Oostendorp (2005), the moraic model introduced ―a popular alternative to the

representations of syllable structure‖ (p. 1). Moreover, Crystal (2008) affirms that the


16

―Moraic structure accounts for many of the phenomena described in other models‖

(p. 312).

The term ―Mora‖ was introduced by Hyman (1984, 1985), McCarthy and Prince

(1986) (McCarthy & Prince, 1990, p. 4). Noske (1993) refers to Mora as: ―the

element which expresses phonological weight (and) plays a major role in syllable

structure‖ (p. 31). Watson (2002) declares that the model ―could adequately reflect

syllables‘ weights‖, as it ―develops a prosodic conception of the syllable, giving

status only to those segments which bear weight and are potentially stressable‖

(p.53). Crystal (2008) elucidates that the ―analysis of segments into moras is usually

applied only to the syllabic nucleus and coda (the rhyme), and not to the onset‖ (p.

312). However, as far as EA is concerned, not all codas are assigned a mora where

this would depend on its location in a syllable. This is referred to as ―a weight by

position‖, which Watson (2002) explaines as: ―a ‗coda‘ consonant is assigned a mora

in the course of syllabification (parametric)‖ (p. 54). It means that codas would count

moraic only after word syllabification, where coda weight would be determined

according to its position, and parametric stands for the language specific nature of

such a principle. This could be deduced from Kiparsky (2003) through the diagrams

below:

(p. 157)

Therefore, the diagrams designate the places of mora, which in turn marks weight. It

reflects the association existing between the number of vowels and the locations of

consonants after syllabification (i.e. 1 vowel=1 mora, 2 vowels = 2 moras, onset= 0

mora, coda in a heavy syllable CVC=1 mora, coda in an ultimate syllable (word

final) CVCVC (extrametrical) = 0 mora, coda in a final position of super heavy

CVVC= 0 mora (extrasyllabic), or CVCC = 0 mora (extrasyllabic)) (Watson, pp. 56-

58).

3. Data Analysis

3.1. The Case of Glottal Stop Alternations in Early EA

Before starting the analysis, an idea is given about the historical background of

the glottal stop.

3.1.1. A Sociohistorical Linguistic Perspective of Al-Alif and Al-Hamza


17

The history of the glottal stop could be traced back through time to Egyptian

and Semitic languages. Kramer (n.d) explicates that the ―glottal stop remained

without a letter in most Coptic dialects, although the distribution is traditionally

constrained to the beginnings and ends of vowel-initial and vowel-final words, and

between doubly written vowels‖ (p. 19). As for Semitic languages, the stop persisted,

where, in addition to Arabic, it existed in Hebrew, Assyrian, Aramaic, and

Abyssinian. In addition, it manifested various phonological alternations in almost all

of these languages, just like the case in Arabic. Moreover, the orthographic history of

the glottal stop in Arabic played a crucial role in its phonological realization across

Arabic dialects (De Lacy, 1923, pp. 30-41).

Badawi, Carter & Gully (2004) assert that ―the original Arabic alphabet had

used the letter ا (Alif)‖, where it stood for both the long vowel /aa/ and the glottal

stop /ʔ/ (p. 1). According to Abdil-Tawwaab (1996), the problem was solved by Al-

Farāāhiidiy, where he introduced a new symbol (ء), thus distinguishing between the

/aa/ and the glottal stop, instead of the common (ا). Moreover, many early Arab

grammarians such as Ibn Ya‗iish and Sibawayh agreed on the marked articulatory

effort that accompanies the glottal stop‘s production. This consequently led to the

wide alternations observed to accompany this phoneme. This was evident through

the dialect of the people of Hiʤaz, particularly Quraysh tribe, who resorted to

forsaking glottal stops in favor of either deletion or alternation to another phoneme

(pp. 24-27). Conversly, some other tribes preserved the glottal stop such as Tamim,

Qays, and Banu Osd, as asserted by Hilaal (1993, p. 220).

In contemporary EA, /ʔ/ is majorly used as a variant of /q/, yet some words

preserved the /q/. Ezzat (1974) points out that the /q/ is still present in Egyptian

Arabic, as in the words /ʔalqāāhirāh/ 'Cairo', and /ʔilquds/ 'Jerusalem'. Moreover, it

could be discerned "when the speaker classicizes or elevates his style" as in /tāqdi:r/

'estimation', and /fāqāT/ 'only' (p.22).

It is to be noted that Al-Zāher Beibers orthogrophically recognized glottal stops

at some instances, which may indicate that it could have been left out at the other

instances to conform to the colloquial dialect used then, as in: , (p. 1518)

/rāʔayta(u)k/ (I saw you), /yaʔxuTHu/ (they took) vs. /shaanha/ (her business) (p.

1521), and /rāāsuh/ (his head) (p. 1764).

3.1.2. Post-vocalic Glottal Stop Elision: Final Position after a Long Vowel The following group is divided into five paradigms, where the division is based

on the word category. All the paradigms encompass words with glottal stop deletion

after a long vowel. For convenience, the examined EA forms are presented using

current articulation (based on their orthography), since it is not possible to present

exact Mamluk prounciation. Moreover, variants, if any, are given between brackets.


18

Table 1

(For more examples, check appendix 1 no. (a))

All the previous data share the same phonetic environment for the glottal stop,

being underlyingly preceded by a long vowel in an ultimate syllable. Onsets do not

seem to confine to a specific place or manner of articulation, i.e. dento-alveolar, trill

/r/, bilabial, nasal /m/, alveo-dental, stop /d/, alveo-palatal, affricate /ʤ/, etc. Hence,

ultimate closed syllables are made up of the templatic structure [CVVC] in

polysyllabic words (i.e. with a disyllabic majority). Glottal stops are observed for

being extrasyllabic in a super heavy, bimoraic syllable all along. Further, the pre-

glottal stop vowel (i.e. bimoraic) has /aa/ as noticeablely dominating (26 out of 38

words) (please check table (a) in appendix 1). These final glottal stops, however, were

elided, leaving a final open [CVV] syllable.

Page no. &

Gloss.

Original form

in CA

EA word

transcribed Word in EA

a) Singular Nouns

(p.1521) destiny /ʔal-qāDāāʔ/اىقضبء /ʔi (a)l-qāDā/ اىقضب

(p.1478) penalty /ʤazaaʔ/خضاء /g(j)(i/a)za)/ خضا

(p.1484) desert /ʔal-

SaHrāāʔ/اىظحشاء /ʔi(a)l-SāHārā/

V insertion اىظحشا

(p.1482) ordeal /ʔal-balaaʔ/اىجالء /ʔi(a)l-bala/ اىجال

(p.1517) night /ʔal-masaaʔ/اىغبء /ʔi(a)l- m(i)(a)sa/ اىغب

b) Plural nouns

(p.1502) guard /ʔal-ghufarāāʔ/اىغفشاء /ʔi(a)l-ghufārā / اىغفشا

(p.1649) enemies /ʔa(i)l-

ʔaʕdaaʔ/األػذاء /ʔi(a)l-ʔaʕda/ االػذا

(p. 1515) poor

people

/fuqārāāʔ/

فقشاء/fuqā(ʔa)rāh/ فقشا

(p.2850) blood /dimaaʔ/دبء /di(a)ma/ دب

c) Verb

(p.1554) I come

/ʔaʤiiʔ/

أخء

/ʔaag(j)i/ آخ

d) Adverb of Place

(p.1501) behind /warāāʔ/ساء /wārā/ سا


19

Birkeland (1952) spells out that ―Cairene does not usually allow word-final

[CVV] syllables as a result of ―the history of the dialect when long final vowels were

reduced to short final vowels‖ (as cited in Watson, 2002, p. 56). The examples above

are assumed to have been subjected to the same shortening process when meeting the

structural conditioning environment, i.e. being located word finally after the glottal

stop elision process. Such historical final vowel shortening has been confirmed by

Watson (2002) stating that CVV does not occur in phonological word-final position.

The following illustration, adopted from Watson (2002), could represent the process:

(p.57)

In addition, the rules given below were deduced to account for the phonological

condition, required for the application of the two processes attested with the set of

words in the table above:

Gadalla (2000) affirms the presence of a regular phonological rule, in which

―[F]inal /ʔ/ Deletion and Final-Vowel Shortening‖ take place in EA (p. 149). The

phenomena are noticed to accompany words of different categories in the data

dispayed in table (1), encompassing nouns (sing. & pl.), adjectives; a verb, and an

adverb of place that occurred repeatedly in the corpus. Accordingly, the final long

vowels become short, thus turning bimoraic, closed syllables into monomoraic, open

syllables. It could be hinted out that postvocalic glottal stop elision might reflect

evidence for being affiliated to the lexical domain as it takes place inside the word, is

1) /ʔ/ deletion /ʔ/ /φ/ / [+long V]---- ] (i.e. glottal stop is deleted in

case it follows a long vowel in a word final position).

2) Final

Vowel

Shortening

(FVS)

[+long V] [-long V] / ------] (i.e. long vowels are

shortened in a word final position).


21

structure preserving, and is non-final being followed by other processes (i.e. reveals

cyclicity) as would be seen below.

Tracing out the phenomenon of glottal stop deletion throughout different epochs,

Ahmed (1993) has reported witnessing it in documents from the Fatimid era,

mentioning words, such as /ʔ(i)al-qāDā/ ضباىق , instead of /ʔ(i)al-qāDāāʔ/, from

Akhbāār Sibawayh Al-MāSri, and /ghada/غذا instead of /ghadaaʔ/غذاء (lunch) from

Manamaatu-l- wahrāāniy (p. 41). As for the Ottoman period, words such as /ʔa(i)l-

rida/ اىشدا for /ʔa(i)l- ridaaʔ/ (p. 16) (gown/robe), and اىغب/ʔa(i)l ghuna/ from /ʔa(i)l-

ghinaaʔ/ (p. 17) (singing) were attested in Hazz Al- QuHuuf by Al-Shirbiiniy (1890). It

was also detected in other Mamluk era documents, such as Al-Bashbaghawiy‘s Nuzhat

Al-Nufuus wa-Mudhik Al-‘Abuus, where it was found to embrace words, such as "عد"

i.e. /suudah/ from ,(as cited in Dos & Davies, 2013, p. 45) "اىا" ,"ثض",

/sawdaaʔ/عداء (black), /beeDāh/ from /bayDāāʔ/ ثضبء (white), and /ʔa(i)l-hawa/ from

/ʔal-hawaaʔ/ اىاء. This, in turn, could prove the antiquity of the phenomenon in EA

tradition.

3.1.2.1. /h/ insertion accompanying final stop /ʔ/ elision

Some peculiar observations could be pinpointed in relation to group (f),

encompassing the following

words,

Group (f) is noticed to encompass words that surfaced with a pharyngeal /h/.

The set appears to share some common factors. First, it includes adjectival words that

exhibit the same CA (UR) templatic measure of [FaʕLaaʔ]. Second, their ultimate

f) Adjectives ( with h/insertion)

(p.1602) black /dahmaaʔ/دبء /dahmah/ دخ

(p.1554) lame /ʕargaaʔ/ػشخبء /ʕarg(j)ah/ ػشخ

(p.1554) one-

eyed

/ʕawrāāʔ/ػساء /ʕoorāh/ ػس

(p.1554) blond /shaqrāāʔشقشاء /shāq(ʔ)rāh/ شقشح

)p.1554) white

mixed

with red or black

/shahbaaʔشجبء /shahbah/ شج


21

syllables, commonly, share the same UR ending, namely, long vowel followed by a

glottal stop /aaʔ/. Hence, as mentioned by Ryding (2005), /aaʔ/ is one of the CA

nominal feminine markers in addition to /aa/, i.e. Alif Tāwiilāā [ا] & Alif MaqSuurāā

[] (p. 123). Converting to EA, by virtue of /ʔ/ deletion and preceding vowel

shortening, the words in (d) developed a coda (/ʕarga<h>/, /shāqrā<h>/, etc.) that is

restricted to the pharyngeal segment /h/. The process could be deduced to be

formulated as such:

The /h/ is regarded here as extrametrical, for according to Watson, ―a word-final

CVC syllable counts as monomoraic‖ (p. 58), being ―prosodically light‖ in such

position (p. 57). Furthermore, /h/ insertion reflected a significant appearance that is,

particularly, associated with the word location in the phrase. It has been observed that

the word /fuqārāāʔ/ was written in Beibers in two ways as: فقشا and فقش (i.e. with and

without the final /h/). The following phrases with the word forms in their contexts

might provide some clues to account for such a phenomenon:

a) /ʔakalna ʔakl fuqārāh / (we ate of the food of poor people) فقش" اميب امو ‖

(p. 1501).

b) / fuqārā-l-balad / (the poor of the country) فقشا اىجيذ "

" (p. 1515).

The word /fuqārā/ in both phrases are noticed to share the same nominal adjectival

category, with the same UR (CA) templatic form [Fuʕalaaʔ], yet the crucial difference

lies in the position of each word. Hence, /fuqārāh/ is located in a final position,

whereas /fuqārā/ is located phrase initially. Hence, the position of the word might be a

triggering factor for /h/-insertion, as it is noticed to accompany adjectives positioned

phrase-finally. However, this postulation is based on the data attested here as extracted

from the epic and requires further evidence to be generalized to EA.

/h/-insertion has its origin in CA phonological tradition. According to Al -Pop

(2010), Sibawayh has termed finally-located inserted /h/ as ―haa‘ al- sakt‖. It denotes

an inserted /h/ that is pronounced in phrase final position or pause ―Al-Wāqf‖. The

/h/ in such location, typically, helps in revealing the ―Harāka” (inflection/diacritic)

of the final segment. The augmented consonant is well-recognized also in Qur‘anic

spelling and recitation, where it can be witnessed with defective imperatives such as


22

/ʔiqtādi(h)/ اقزذ (follow as an example), and the noun /sulTāāniya(h)/ عيطب (my

power/ my sultanate), etc. (p. 21).

/h/ insertion, consequently, could be classified as being located in the

postlexical module, because it applies after the phonological phrase level rather than

the word level. This means that the epenthesis occured after the clustering of words,

and syllabification. It takes place in regard to the location of words in a phrase.

Moreover, it may be sensitive to speed and register. These reasons, therefore, could

support the postulation that /h/ insertion takes place in the postlexical module. The

phenomenon could also be regarded as a supporting evidence (i.e. along with the

synchronic evidence) that corroborates for final vowel shortening, as none of the

words with /h/ insertion surfaced with a pre-/h/ long vowel.

3.1.3. Post-Vocalic Glottal Stop Elision in CVC Syllables

The original forms (CA) in the following table display an ultimate syllable

structure of (CVC) with a glottal stop that is located post-vocalically. The table is

divided into two paradigms based on the category of each set, where CA and EA are

to be examined. It is to be noted that the punctuation of EA words is added by the

researcher, while CA words are taken from baheth.net. Moreover, roots are added

next to the derived words for clarification.

Table 2

Word in EA EA word

transcribed

Original

form in CA Page no. &

Gloss.

a )Verbs

ازؤل/ʔimtala/ /ʔimtalaʔ/ ازال اىغي[m,l,ʔ]

(p.1600) to get full

(p.1612) [w,k,ʔ] ارنأ/ʔittaka/ /ʔittakaʔ/ ارن

leaned yi(a)t(d)daffa/ (V/ زذف

elision)

/yatadaffaʔ/زذف [d,f,ʔ]

(p.1501) to get

warm /yithanna/ ز

(gemination

of /n/)

/yahnaʔ/أ [h,n,ʔ] (to be pleased)

/t(T)āwāDā/ رض(change in V

quality)

/tawaDDāʔ/أ ض ر[w,D,ʔ]

(p.1576) performs ablution

-/tinTifi/ رطف

/ti(a)nTāfa/

/tanTāfiʔ/رطف [T,f,ʔ]

(p.1480) to (be) put

out


23

b) Nouns

قش /muq(ʔ)ri/ /muqriʔ/قشئ [q,r,ʔ] (p.1816)

reciter/reader اط اقذال /māwāāTi/ /mawāāTiʔ/اطئ

[w.T.ʔ]

(p.1532)

foothold

(For more examples, check appendix 1 no.(b))

Examining the data above, it could be observed that all stems are derived from a

set of roots that exhibited the same structure of [C. C. ʔ], i.e. [F.ʔ.L] ‗d.f. ʔ‘, ‗q.r. ʔ‘,

‗m.l. ʔ, etc.‘, as checked by the researcher in Elixir and baheth.net. They noticeably

have in common the glottal stop functioning as the third radical, namely, the laam of

the verb on the templatic pattern [F.ʔ.L]. Second, they range between perfective and

imperfective verbs, nouns and adjectives, where they are all derived from the same

triconsonantal root structure. Each corresponding colloquial form shows a systematic

loss of the coda of the ultimate syllable, typically with no further changes to follow, as

detected in the data. The rule operating could, consequently, be formulated as given

below:

/ ʔ / φ / [- long V] --- ]

(i.e. glottal stop is deleted when preceded by a short vowel before a word

boundary).

The final glottal stop elision rule here appears to be similar to that applied to the one

deleting glottal stop after a long vowel [CVVʔ] in table 1. This would lead to the

following rule combining the two cases:

/ʔ / φ / [+/- long V] ---- ]

(i.e. glottal stop is deleted when preceded by a short or a long vowel before a

word boundary).

3.1.4. Medial Glottal Stop Elision:

3.1.4.1. At the End of the Lexical Morpheme before Suffixes

The data below manifest a set of words that surfaced with no glottal stop,

i.e. in their final forms. However, by inspecting CA versions, it was noticed that

each word has a medial glottal stop that is precisely located at the end of the

lexical morpheme, before the beginning of the attached suffix.


24

Table 3

Words displayed in the above table are all suffixed to a personal pronoun and/or

a particle. A characteristic feature of these lexical units is their CA underlyingly

terminal glottal stop (before the suffixes), which follows a long vowel all along. This

in fact is noticed to be equivalent to the data in table (1), which, predominantly,

exhibited the same UR structure of an ultimate super heavy syllable in nominal and

verbal stems.

The data above are classified into two groups; group (a) includes a set of nouns,

whereas group (b) encompasses perf., and imperf. verbs. The surfaced short vowel-

ending-lexical morphemes, i.e. /ghada/, /ʔāʕDā/, after undergoing glottal stop

omission & VS, are noticed to attach to five kinds of suffixes. Suffixes detected are:

1st pers. plu. [na], 2

nd pers., sing., masc. [ak], realized here as [k]) (you). According to

Page no. &

Gloss. Original form in CA EA word

transcribed Word in

EA

a) Nouns

(p.1497) your

lunch

/ghadaaʔak/غذائل /ghadaak/ غذاك

(p.2846) your

friends

/rufaqāāʔak سفقبئل /ru(i)faqāāq/

currently /rifaaʔ(q)/

سفقبك

(p.1648) your

organs /ʔaʕDāāʔak/اػضبئل /ʔāʕDāāk/ اػضبك

(p.1608) his

penalty

/ʤazaaʔah/خضاء /g(j)azaah/ خضا

(p.1592)meeting-

him

/liqāāʔih/ىقبئ /lu(i)qāā(ʔaa)h/ ىقب

(p.1514)keeping

it/her

/ʔibqāāʔi(a)ha/إثقبئب /ʔi(a)bqāāha/ اثقبب

b) Verbs

(p.1608)

to

bring to you

(S&pl.)

to come

1-/ʤaaʔa b-/ (blend) خبء ة /ʔaʤiiʔ(u) lak(um) b-

/ʤaaʔana/ أخئ ىن ة -خبء

or

2-/ ʤaʔab/خأة

/ʔag(j)iib lak/ /gab lu(a)kum//

/gaana/

/ʔagii lak/

اخت ىل خبة ىن

خبب أخيل


25

Watson (2002), [k] occurs post vocalically e.g. /ghadaa.k/, whereas [ak] occurs post

consonantly, e.g. /ʔakl.ak/ (p.181). There are also the 3rd

pers. masc. sing. [-h], and the

prepositions [li-] and [bi-], attached to /ʤaaʔa/. Abdil -Tawwaab (1997) attributed the

output /ʤaab/ to combining /ʤaaʔa/ (came) and /bi-kaTHa/ (with something) (p. 145).

However, it has to be mentioned that /ʤaʔaba/ has been traced out in Lisaan Al-Ārāb

with the meaning ‗to earn‘ خأة دأة خأثب: مغت , which could be regarded as the word‘s

origin, yet with undergoing slight semantic shift, as it means here ‗to bring‘.

Moreover, /ʤaaba/ meaning ‗to answer‘ ة اإلخبثخ أ أعشػ إخبثخ، خبة has also been أخ

detected.

McCarthy (1992) elucidates that there ―are no stem-final vowels that remain

short before a suffix‖ (p. 1). Watson (2002) agrees, stating that the output of such

suffixes cliticization should undergo ―mora reduplication‖ as a reaction to a constraint

forbidding monomoraicity in a pre-suffix position. This is represented as follows:

(p. 202)

In addition, VL in pre-consonantal suffix context should be ordered in a late

stage in the lexical derivation. This is attributed to the fact that /ʔ/ omission is

followed by vowel shortening in prior steps, resulting in the creation of a derived

environment for pre-suffix VL to take place, i.e. cyclicity is an indication of lexical

rules, as explained on p. 12. Watson (2002) also confirms that pre-suffix vowel

lengthening is affiliated to lexical module. This was detected in the data under study;

for the rules have, all through, been witnessed to confine to word boundaries.

Moreover, they had an access to internal word information, and did not result in a

new segment or sequence, thus it is stipulated to take place in the lexical stratum.

Based on the data extracted from the epic, the phenomenon seems to apply

systematically, wherever the phonological conditioning environment is provided.

3.1.4.2. Glottal stop elision in V--- C position

Glottal stops in the next paradigms are noticed to be located in syllable coda

position in CA stems (UR).The data below are classified into two paradigms based on

the environment of the glottal stop. Moreover, roots are added next to the derived

words for clarification.


26

Table 4

Page no. & Gloss Original form in

CA

EA word

transcribed Word in

EA

a) glottal stop elision in V_ C position

1-Verbs

(p.1700) you

deserve

/tastaʔhil/

[ʔ. h.l]رغزأو) )اعزأي اعزخج

/ta(i)staahil/

رغزبو

(p.1501) wait for me /ʔistaʔnuuni/اعزأ [ʔ.n.y]

/ʔistannuuni/

(sometimes

اعز


27

The glottal stop is, generally, noticed for being a member of the radicals of each

of the above words (except for /ziʔbāq/), whether verbs or nouns, i.e. (ʔ. h. l), (ʔ.

x.TH), (k. ʔ. s), etc. It is located in CA stems as occupying the coda position in either

a non-ultimate, heavy syllable (CVC) of a polysyllabic word (i.e. verbs in the table

above), or in a super heavy (CVCC) of a monosyllabic word (i.e. nouns in the table

above). In addition, glottal stop elisions take place with all word categories, i.e. verbs

with different inflections and tenses, nouns, and cliticized forms. Arabicized words,

such as /ziʔbāq/, i.e. arabicized from Persian (baheth: Lisaan-l-‘Ārāb), could be also

affiliated to the same group as they reflect the same change conditioned by the

phonological environment. Thus, all words have undergone medial glottal stop

elision, as a consequence of the following deduced rule:

Coda

spotted currently

as /ʔistaʔana/ ) (p.1501) to take

(plu.)

/(ya)(na)ʔxuTH.haa/أخزب [ʔ. x.TH ]

/(ya)(na)xuTH(z)(ha)/ بخز- ب((بخز

(pp.1597-1612

to eat)

/(n-)(t-)(y-)aʔkul/أمو [ʔ.k.l]

/naakul – /taakul/-/yaakul/ بمو-

بمو-ربمو

2-Nouns

(p.1480)

cup

/kaʔs/مأط [k.ʔ.s] /kaas/ مبط

(pp.1488, 1505)

well

/biʔr/ثئش [b.ʔ.r] /biir/ ثش

(p.1510) axe /faʔs/فأط [f.ʔ.s] /faas/ فبط

(p.1698) revenge /thāʔr/ثأس [th.ʔ.r] /thāār/ currently as

/(T/t)āār/ ثبس

(p.1510)head / rāʔs /سأط [r.ʔ.s] /rāās/ ساط

(pp.1505, 1577)

for him

/ʕalaa shaʔnuh/ شأ ػي [sh.ʔ.n]

/ʕalaa shaanuh/

now cliticized as

/ʕalashanuh/

ػي شب

3-Arabized word (p.1702)mercury /ziʔbāq/صئجق /zeebāq/(aʔ)/

(Arabacized from

Persian)

صجق


28

/ʔ/ φ / V C

(a glottal stop is elided in coda position in case it is immediately preceded

by a vowel, and immediately followed by a consonant).

This fact has been affirmed by Gadalla (2000) who states that ―[T]he glottal stop

is ... lost in EA when it forms all or part of the coda of a syllable...‖, which leads to

―compensatory lengthening‖ of the adjacent vowel (p. 30). This conforms to the data

here, where a repairing process of vowel lengthening was noticed to operate to

balance the guttural segment loss and to preserve the UR syllable weight (bimoraic).

The phenomenon is described by Crystal (2008) as ―an effect in which the deletion

of one segment is accompanied by an increase in the length of another usually

adjacent to it, thus preserving the syllable weight‖ (pp. 91, 92). Similarily, a stability

in the syllables‘ weight is noticed through the examples above after undergoing

glottal stop deletion (i.e. CVʔ <CVV, CVʔC < CVVC). Watson (2002) explicates

that the deletion would cause freedom of the ―right hand mora‖, represented by the

vowel, leading to its ―spread‖ or in other words lengthening. She further illustrates

the process as follows:

(p. 55)

The same phenomenon seems to have taken place with the word /ʔistaʔnuuni/

which according to Abdil Tawwaab (2007) is the source of the colloquial ,اعزأ

/ʔistannuuni/ (p. 396). The word was looked up in baheth: Lisaanu-l-Ārāb, where it

was found as denoting the same meaning intended here: ― ف أشك أ ال قبه: اعزأ

ذ ثن أ ازظشد رشثظذ It is said slow down or do not hasten, also I am) .‖رؼدو، اعزأ

waiting for you, which means that I awaited and expected). The word must have first

passed by a glottal stop deletion, i.e. /ʔistanuuni/. Nevertheless, it could be deduced

of the word‘s spelling اعز that, instead of a compensatory vowel lengthening, the

adjacent /n/ might have become geminated, hence, /ʔistannuuni/ (consonant

lengthening), which is the form attested at the present time. However, the gemination

cannot be totally affirmed during that time due to the absence of diacritics in the

epic. In addition, n-strengthening was concluded to be related to the lexical domain

as confirmed by Watson (2002, p 205). It could be argued therefore, that signs of

cyclicity could be reflected through the glottal stop deletion followed by the vowel

lengthening. Some words are noticed to escape the deletion process nowadays. This

is seen in synchronic lexical exceptions, such as /maʔzuun/ (marriage official) أر ,

/miʔzanah/ ئزخ (minaret), /maʔsaah/ أعب (misery). Moreover, the deletion process


29

did not result in novel structures (i.e. templatic forms). All these proofs signify the

lexicality of the process.

3.1.4.3. Glottal Stop Elision in V---V Position

The second paradigm manifests the omission of the medial glottal stop between

two vowels, as could be revealed through the data below:

Table 5

In group (b), the glottal stop is located between two short vowels. All words are

noticed to surface without the glottal stop; yet, no further vowel lengthening is

detected. This could be attributed to the fact that the deletion process resulted in

concatenation of vowels, thus, leading automatically to long vowels. It is worth

noting that the process here would cause a syllable subtraction, where instead of

/saʔal/ consisting of /sa/, weiging one mora and /ʔal/ weiging one mora, the whole

word became bimoraic /saal/.

Abdil-Tawwaab (1996) confirms glottal stop disappearance between identical

and even non-identical short vowels in some Arabic dialects. He mentioned the words

/saal/ instead of /saʔal/, and /suil/ ه ع instead of /suʔila/ عؤه (was asked), thus leading

to ―Hiatus‖ phenomenon (p.28), or the adjacency of two vowels that belong to

different syllables (Crystal, 2008, p. 228). Nevertheless, the glottal stop in such

position is retrieved in some dialects, i.e. Cairene, e.g. /kaʔinnak/ مأل, /suʔaal/ عؤاه

(question), and omitted in others, e.g. ‗kannak ya-‘Abu Zeed ma ghazeet‘ (as if you,

abu zeed, have not invaded…).

Page no. &

Gloss.

Original form in

CA

EA word

transcribed Word in

EA

b) glottal stop elision intervocalically

as if you /kaʔannak/ ل مأ

/kaa)n)nak/ مبل

(p.1605) to ask /saʔalt/عأىذ (also imp. /sal/ عو)

/sa(a)lt/ عبىذ

(p.1582) to see /rāʔa/سأ /rāāy(a)/ سا


31

To conclude this section, the two instances of glottal stop medial deletion occur

within the word boundary, and the deletion process did not result in novel structures

(i.e. templatic forms). There are also signs of cyclicity, which is reflected in the

deletion, which is followed by the vowel lengthening. In addition, the word

/ʔistannuni/ might provide an evidence of cyclicity, (in case of /n/ gemination), as the

stop deletion is followed by the doubling of /n/, where ―/n/ strengthening‖ before

vowel initial suffixes (in our case (/u/) has been classified as lexical by Watson

(2002, p. 205). The words in group (a) have lexical exceptions that could be viewed

in synchronic data in such words as /maʔzuun/. As for group (b), the rule seems to

have ceased in some words such as /saʔalt/ and /kaʔannak/. All these proofs might

support the postulation that medial stop deletion pertains to the lexical domain.

It is worth noting that one of the seven avowed Qur‘anic recitations, namely

Warsh ‘an Naafi’, is marked by glottal stop deletions, e.g. /yuuminuun/ instead

of /yuʔminuun/ؤ (have faith), e.g. /biir/ثش instead of /biʔr/ ثئش (well), /biis/

instead of /biʔs/ثئظ (misfortune) (MaSri, 2009, p. 28). Moreover, the phenomenon of

medial glottal stop elision (i.e. in both environments V__V & V__C) has been traced

in the Fatimid era through words, such as "اشار" (Ahmed, 2104, p. 95) /ʔimrāātuh/

for /ʔimrāʔatuhu/اشأر(his wife). Furthermore, Aboul-Enein (2012) mentioned the

words /rāās/ and /faas/ from the 17th century (pp. 27, 29).

3.1.6. Initial Glottal Stop Epenthesis

Generally speaking, the phenomenon of inserting a glottal stop in a word initial

position is widely used in Arabic. It is known in CA as Alifu-l-WaSl (junctural

Hamza), and it is described by Ryding (2005) as ―a phonetic device affixed to the

beginning of a word for ease of pronunciation‖ (p.19). Sibawayh (1982) defines it as

an augmented element that is used in case a word begins with a ―saakin‖ that would

result in two consecutive consonants (p. 144). Broselow (1976), McCarthy (2007) and

Badawi et al. (2004), amongst others, agree that the glottal stop insertion is preceded

by another intermediate level. In such a level, vowel epenthesis takes place, where the

glottal stop insertion follows to prevent starting with a vowel, which is another

phototactically disallowed condition in Arabic. Badawi et al. (2004) explain that

―*CCVC becomes VC-CVC, but since no syllable can begin with a vowel, a Hamza

is pronounced‖ in CA. In addition, the term Alifu-l-WaSl meaning the

joining/connecting Alif could perfectly denote its sensitivity to its phrasal location,

henceforth postlexical, as it would solely be pronounced in a word as well as in a

phrase initial position, e.g. /ʔuktub/ امزت (write) (imp.) (p. 12). This was also

confirmed by Watson (2002) as she states that in Cairene ―the glottal stop acts as the


31

postlexical default consonant: this is the consonant which is inserted in the default

case to provide an onset for an otherwise vowel-initial syllable‖ (p.38).The process

could be formulated through the rules assumed below:

1) φ V / [ __ CC

2) φ /ʔ/ / [ __ VCC

Grammarians such as Sibawayh and Ibn Ginni specified certain word categories

as well as certain strategies for inspecting Hamzatu-l-WaSl as opposed to Hamzatu-l-

QāT’. Ibn Ginni (1993) asserted that, generally, Hamzatu-l -WaSl is common amongst

verbs rather than nouns or particles. He elucidated that it comes initially with perf.

verbs, in case the verb exceeded four letters like /ʔiHmārrā/ احش (turned red),

/ʔistaxrāʤa/ اعزخشج (extracted). He added that it also accompanies imperative verbs,

provided that they have their imperf. prefix [ya] followed by a saakin letter [i.e. CC],

hence, /yāDrib/ (imperf.), /yashrāb/ (p. 111). The imp. form of such imperf. verbs

should be attached to Hamzatu-l-WaSl, e.g. /ʔiDrib/ اضشة (hit) & /ʔishrāb/ اششة

(drink). Concerning EA verbs, they seem to conform to the same rule of Hamzatu-l-

WaSl initial epenthesis. This could be concluded from observing the paradigm below,

where the punctuation for the EA is provided by the researcher:

Table 6

Group (a) includes imperative verbs, where the EA forms are noticed to have an

initial syllable of a glottal stop followed by a vowel. This appears to be similar to the

case of Hamzatu-l-WaSl, which might be enhanced by considering the imperf. forms

of the imp. The imperf. forms in this group are supposed to be initiated by the

structure [ya+CV_] in CA, where Hamzatu -l-WaSl is not required in such a case, i.e.

/yu.naadi/, /ya.qif/, and /ya.tasalla/. Nevertheless, Hamzatu -l-WaSl annexation here

might be attributed to deriving EA imp. forms from EA imperf. counterparts, i.e.

Page no. &

Gloss.

Original form

in CA

Word in

EA EA word

transcribed

a) Imperative

(pp.1503,

1514) call

/naad/بد imperf.: /yunaadi/بد

ذ /ʔindah/ ا

(p.1596) have

fun

/tasal(la)/رغو imperf.: /yatasalla/

/ʔitsalla/ ارغي

(p.1586) stop /qif/قف imperf./yaqif/قف

/ʔuqā(ʔa)f/ ا قف


32

[ya+CC_ ] structure (i.e. the colloquial imperf. prefix might also be assumed to be

[yi] or [yu]), i.e. /y(a)(i)ndah/ and /y(a)(i)tsalla/, where these forms would match the

requirement for Hamzatu -l-WaSl insertion. As for the verb /ʔuqaf/, it might have

been derived from the colloquial imperf. /y(a)(i)wqā(ʔa)f/ (common in the Levant).

The following table displays a paradigm of perf. verbs that also show an

insertion of an initial syllable consisting of the [ʔV] structure. The verbs are

mentioned in their contexts to affirm their past tense. Though this case looks similar

to the previous condition of imperative verbs, yet it shows distinctive differences

from the state above. This could be affirmed via investigating the data below:

Table 7

Table (7) encompasses EA verb forms that are noticed to agree with the

templatic form IV in CA or [ʔaFʕala]. According to Ryding (2005), this form could

be, generally, compared to form I or [FaʕaLa], where the form IV is regarded as a

transitive version giving a causative meaning e.g. [ʕalima] ػي (to know) (I) vs.

[ʔaʕalama] أػي (IV) (to inform) (p. 516). However, adding [ʔV] to perf. triliteral

verbs cannot take place all along, as some forms would result in either a different

meaning or else turn to a form that is deviating from CA. Thus, on searching for the

set of EA verbs in the paradigm above in Elixir, and in baheth.net, it was affirmed

that the forms /ʔakaada/, /ʔaHka/do not exist in CA, whereas /ʔāDmānuh/ (I

guarantee him) was found, yet in the 1st pers. sing. imperf. form ذ اىشء أض ض

Therefore, it is evident that the perf. forms attested here belong to the .ضبب

colloquial variety, rather than to CA. Hence, the process involved in the conversion

from CA to EA was accomplished via adding the glottal stop and the instantly

following vowel, resulting in the perf. [ʔaFʔaL].

It should be pointed out that the systematic attitude of Alif Al-WaSl is,

particularly, associated with CA. It is transferrable to borrowed words, i.e. a feature

Page no. &

Gloss.

Original form in

CA

EA word


b) Perfective verbs

(p.1500) was

about to

/kaadat(a)/مبدد /ʔaka(a)dat/ ا رفطش شاسر فأكادت

(p. 1587)

narrated /Hakiya/حن /ʔaHka/ ا اىيل ػشص لو فاحكا …

(p.1617)

guaranteed him

/Dāmantuhu/ زخ مب اضمنتو اذ ارا /ʔāDmāntuh/ ض …خب

( p.1638) arrested /qābāD(ā)/قجض /ʔāqbāD/ ػي ىذ اث االبفاقثض


33

of postlexicality, which could be witnessed in the adaptation of even nouns starting

with initial clusters such as /stadium/ < /ʔistaad/. Thus, it is productive. Moreover,

the process could be specified as postlexical as it takes place in respect to the

position of the verb in a phrase, and appears to be phonollogically conditioned. As

for the case of glottal stop initial epenthesis in group (b), it seems to have taken place

lexically, as it did not show to exceed the word boundary. However, the verbs here

are noticed to be preceded by the causative [f] and the conditional [ʔiTHa], which

might have triggered the stop epenthesis, where then it would be postlexical (table 7).

Moreover, to the knowledge of the researcher, the phenomenon became archaic as it

is not commonly attested synchronically, which makes it exclusive to that particular

time.

3.1.7. /ʔ/ Alternation to the Glide /y/

3.1.7.1. Singular and plural nouns

CA words within the table below are all noticed to surface with altered glottal

stops. This time the stop is located in syllable onset position, in a word medial

position (sing. & plu. nouns). The behavior of glottal stop alternation is inspected

from the data below:

Table 8

Word in EA EA word transcribed Original form in CA Page no. &

&Gloss.

a) Masc. Sing. Nouns [FaaʕiL]

غبت/ghaa(āā).yib/

/ghāā.ʔib(un)/ غبئت (p.1623)

absent

(p.1505, 1717) خبئ/xaa(āā)y in/ /xāā.ʔin/ خب

traitor

(p.1496) غبئض/ghāāy.iS/ /ghāā.ʔiS/ غبض

Plunger

/faayir/ /faaʔir/ فبش فبئش (p.1586)

(effervesce) inquirer (p.1503) عبئو/saa.yil/ /saa.ʔil/ عبو

(p.1642) حبئش/Haa.yir/ /Haa.ʔir/ حبش

confused

(p.1500) الئق/laa(āā).yiq(ʔ)/ /lāāʔiq/ الق

suiting

b) Fem Sing. Noun [FaaʕiLah]


34

ساحخ

/rāā(aa).yHah/

/rāā.ʔi.Hah/سائحخ (p.1502)

smell

(For more examples, check appendix no. (c))

Words in groups (a) and (b) are observed to display the templatic measure of

[FaaʕiL]. This is the templatic form of active participles (APs) derived from three

radical-rooted verb being checked in Elixir. APs are well- recognized for

semantically referring to agentive roles, such as attested in /kataba/-/kaatib/. The

Glottal stop in CA word is consistently located in the ʕiin position of [FaaʕiL]. Thus,

it was observed to regularly occupy the onset position of the 2nd

heavy syllable.

Furthermore, the AP forms are commonly derived from hollow verbs, where

according to Ryding (2005), these verbs have semi- consonants (glides) or as the

middle radical (p. 461). Hence, glides are observed in the mid of the root of each of

the words above, i.e. [f. w. r], [x. w. n], [H. y. r], etc. Broselow (1976) affirms that

―participles of hollow verbs (in EA) have...the glide which appears on the

surface...(an) invariable /y/, even in verbs with putative underlying /w/‖ e.g.

―/naayim/‖, ―/xaayif/‖, ―/shaayil/‖ (p.168), as [n.w.m], [x.w.f], and [sh.w.l]. Hence,

the glide /y/ proves to be a consistent alternant of the glottal stop in AP words, which

are derived from hollow rooted verbs.

Gadalla (2000) states that ―the change of the glottal stop /ʔ/ to the high glide /y/‖

takes place ―when it is followed by /i/ and preceded by any vowel‖, presenting the

formula ―Vʔi → Vyi‖ (p. 33). Thence, the regularity in the data attested here might

result in the formulation below, (i.e. the rule is restricted to the set of words exhibited

above).

+stop +high - high V - low V

+glottal +glide / +long V ___ - long V

+front V

(i.e. glottal stop turns to a high glide in case it is preceded by a low,

long vowel and followed by a high, short, front vowel )

The rule has proved to apply to words showing the same phonological

environment though belonging to various sub-categories (as would be also noticed in

table 9). This could be clearly demonstrated by considering the CA word /rāāʔiHah/

(i.e. homophonous with the word denoting ‗she is going‘), which is classified as a

fem., sing. noun. The word designates that even though it does not belong to the

broad morphological category of APs, yet it underwent the same process of stop

conversion, because of being measured against the templatic form [FaaʕiL]. To the


35

knowledge of the researcher, the word /rāā)aa)yHah/ساح, meaning smell, is not used

as such nowadays, where the diphthong /ay/ turned to the long vowel /ii/, giving rise

to /riiHah/.

3.1.7.2. Plural nouns

The glottal stop alternation to /y/ was to affect the following nouns as follows:

Table 9

Word in EA EA word

transcribed Original form

in CA

Page no. & &Gloss.

c) Broken Plural Nouns/FaʕaaʔiL/

pillows عبئذ/wa.saayid /wa.saa.ʔid/ عبذ رفشقؼذ

المراير/mā.rāā.yir/ /ma.rāā.ʔir/شائش (p.1834) gall

bladder (p.1800( نبئذ/ma.kaa.yid/ /ma.kaa.ʔid/ نبذ

intrigues conscience (p.1652) عشائش/sā.rāā.yir/ /sa.rāā.ʔir/ عشاش

good traits (p.1633) شبئو/sha.maa.yil/ /shamaaʔil/ شبو

Table (9) encompasses another group that belongs to the same broad class of

nouns, i.e. broken plural nouns. Moreover, pluralized forms agree in having singular

alternatives that are noticed to range between the templatic form [FaʕiiLa(h)], e.g.

/sariirā(h)/ and [Fiʕaala(h)], e.g. /wisaadah/. The sing. skeletal form [CVCVVCa(h)],

in general, has been referred to by Ryding (2005) as characterized by an insertion of a

medial or initial glottal stop in the plural formation (p. 152). Accordingly, by

investigating the words in (c), it could be observed that the sing. forms, of the above

mentioned forms of EA, are fem. where they all have the [a(h)] marker along with

having long vowels in the penult. syllable, e.g. /marāārā(h)/, /makiida(h)/, etc.

Henceforth, they commonly reflected an insertion of glottal stop word medially in the

plu. CA form, whereas in EA plu. version of the data, the glottal stop was missing.


36

This came as a result of following the same previous postulated rule of glottal stop

alternation into the glide /y/.

3.1.7.3. C-----[+long] Adjectives

Investigating EA words in table (10), the glide /y/ has also been detected as an

alternative to the glottal stop in CA words.

Table 10

Word in EA EA word

transcribed Original form in CA Page no. & &Gloss.

e) C___ [+long V]

full (p.1574) آل/mal.yaan/ /mal.ʔaan/ يب /may.shuum/ /mash.ʔuum(ah)/ ش

شؤ

(p. 1590) ominous

Similar to the previous groups, the glottal stop here is noticed to occupy an

onset position in word ultimate syllables. The main difference, however, lies in the

surrounding environment, where the stop here is preceded by a consonant and

followed by a long vowel. This might permit the following formulation to be stated,

as specific to the data:

+stop +high

+glottal +glide / + C -short V

(The glottal stop turns into the glide /y/ in case it is surrounded by a consonant and

followed by a long vowel)

Glottal stop alternation in the data above did not surpass the word boundary. Looking

up the word /malaʔ/ in Lisaan Al-Ārāb, it was found that there are other forms than

/malʔaan/ like ؤل آلخ …إبء آل) (a full vessel /malʔaan/, there is also the masc.

/malʔan/ ؤل and the fem. /malʔaanah/. Hence, it is worth noting that the form /malaan/

with the glottal stop omitted was mentioned in Lisaan Al- Ārāb as well,


37

إ شئذ خففذ اىضح اىؼبخ رقه : إبء ال ف اىزمش ال" ‖ ، (Hamza could be alleviated,

folks say ‗full vessel /malaa/‘, and in the masc. ‗/malaan/‘), which could provide

evidence that enhances the postulation of glottal stop alternation being originally

transferred into EA through the dialects of some Arabic tribes. According to Ahmed

(2014), the word /mayshuum/ has been witnessed in Fatimid era documents, such as

Manamaat Al-Wahrāāniy, i.e. ―ش اىطيؼخ ف مو ح ‖ (and he is accursed at all

times). He deduced that the word has first become /mashyuum/, and was affected by

metathesis, which consequently led to the form /mayshuum/ (p.132). In addition, the

word /mayshuum/ seems to have undergone a further glide omission turning to

/mashuum/ as in folkoric proverbs, mentioned by Salah El-Din (2010): أثك اىجظو "

,Your father is onion and your mother is garlic) أل اىز، ىل اىشحخ اىطجخ ب ش "

from where would you get the good smell, you cursed).

The alternation process attested in tables (8, 9, and 10) could be affiliated to the

lexical stratum, as it shows no signs of being applied across word boundary.

Moreover, the segment resulting of the alternation process is a member of EA

phonemic inventory. It also shows to have exceptions which could be revealed through

synchronic data such as /mutafaaʔil/ زفبئو (optimistic), /mutashaaʔim/

which have the glottal stop located in a ,(problems)غبئو /masaaʔil/,(pessemestic)زشبئ

similar environment of an /aa/—/i/ sequence to those in (a &b), and /masʔuul/

which has the C—[+long V] sequence in (e). Accordingly, for ,(responsible)غئه

these reasons the process here could be regarded as lexical.

3.2. Vowel shortening (VS) as a Model of Vowel Alternation

3.2.1. Nouns

In addition to the previous examples of VS, the following paradigms include a

variety of words that display vowel shortening, as could be noticed through the

following table:


38

Table 11

Examining group (a), it could be noticed that the words are originally derived

from three radicals and share the active participle templatic form [FaaʕiL] (Elixir).

Moreover, all the data are suffixed to the fem. marker [a(h)] i.e. [FaaʕiL(a(h))] (i.e.

with the exception of /qāādiriin/, which ends in a plu. 3rd

pers. suffix). The initial

syllables of each word agree on being bimoraic ones, structured as CVV in CA (UR).

According to Broselow (1976), VS takes place right after the application of a high

vowel deletion rule (HVD) (p. 19). The rule deletes the high, front vowel /i/ in EA, in

case the morpheme becomes attached to a pronominal suffix, provided that the

morpheme is vowel-initial, i.e. /ah/, /u/, etc. This gives rise to the adjacency of the

long vowel to a consonant leading to [CVVC.CV--] structure. In addition, the

syllable location in a penult. position would make its final consonant moraic (non-

extrametrical). Accordingly, Vowel shortening operates to prevent the occurrence of

a forbidden trimoraic syllable. Abu-Mansour (1992) expounds that:

In moraic theory, Closed Syllable Shortening is a consequence of the

two mora limit on syllables. A long vowel occupies two moras, but a

syllable final consonant is also moraic. Thus, a long vowel and a

syllable final consonant cannot co-occur, the vowel shortens instead.

Word-final consonants do not trigger Shortening because they are

extrametrical. (p. 53)

Vowel shortening could be observed to reflect a systematic, analogical attitude that is

even persistent till the present time, as in /fāāTimah/ فبطخ < /fāTmāh/ فطخ &

/naaʤiHah/ بخحخ < /nagHah/ دحخ.The rule was formulated by Gadalla (2000) as

follows:

Page no. & Gloss. Original form in

CA

EA word


a)nouns ( p.1503) your

maid slave

/ʤaa.ri.ya.tak/خبسزل /g(j)aryitak/ خشزل

(p.1616) incident

/wāā.qiʕah/اقؼخ

/wāq(aʔ)ʕah/ ب ىب اقؼخوقعة مبذ

(p.1482) crisis

/faa.ʤiʕah/فبخؼخ /fag(j)ʕah/ فجعة

(p.1607) they

can

/qāā.diriin/قبدس /qā(ʔa)driin/ ؼيا قذريهاش


39

1) HVD:

+high V

- long V φ / VC ______ CV

(p.241)

- stress V

A subsequent vowel shortening rule based on trimoraic restriction could be

postulated as such:

2) VS:

[+long V] [-long V] / ___C.CV

(VS takes place, because the final consonant in a super heavy syllable [CVVC] (i.e.

/waaq/) would be moraic (equals 1 mora), resulting in a syllable that weighs 3 moras;

because of being located in a penult. position (not extrasyllabic), leading to VS).

The derivation should then be ordered as follows:

Lexical Module

UR: [wāā.qi.ʕ]

Suffixation: [[waaqi.ʕ][ah]] CA

HVD: [wāāq.ʕāh]

VS: [wāqʕāh] EA

PF (i.e. output): /wāqʕāh/

Watson (2002) has affiliated VS to the postlexical module, while (Abu-

Mansour) linked it to either the lexical or postlexical in EA, considering HVD as

postlexical. However, Berendsen (1986) asserts that HVD could also be lexical

stating ―I have not been able to discover any facts leading to a descision about the

proper position of HVD‖ (p.126). Hence, the domains of such rules are

controversial. Thus, they would be substantiated based on the data under study.

According to the data exhibited here, the process might be assumed to have taken

place in order to prevent trimoraic syllable, primarily to preserve the structure of a

maximum bimoraic permitted syllables. As Borowsky (1989) remarks, VS could,

also, be witnessed in English, which is evident in deriving the past participle /ment/

(meant) from /miin/ (mean) instead of */miint/. This happened as a result of closed

vowel shortening rule that shortens vowels ―when they are followed by a

tautosyllabic consonant (adding) that it is triggered by Structure Preservation‖ (p.

150).

VS, as regards the current data, might be regareded as lexical, as it showed

to follow HVD. This may, therefore, denotes cyclicity and application in a derived

environment, which is defined by Pandey (1997) as ―an environment created by the


41

application of an earlier rule in the same cycle‖ (p.99). In addition, the output is

restricted to a preserved structure, as neither novel segments nor sequences were

created. This might provide evidence for the lexicality of the process (i.e. unlike the

flapping attested in English, where it resulted in the surfacing of the allophone [ɺ]

as in [bu/ɺ/er], which is, according to Mohanan (1986), not part of English

phonemic inventory. Hence, since the structure is preserved, this may be another

proof of the lexicality of the process.

3.2.2. Miscellaneous Categories

The following group also has an underlying long vowel that surfaces as a short

one, as could be deduced from the paradigm:

Table 12

Word in EA EA word transcribed Original form in CA Page no. &

Gloss.

b) Misc. Categories

شح حج ب ... /Hag/ (n.) /Haaʤ/ حبج (p.1514) pilgrim

اث تجنة /bi-gan(m)b/ (prep-

adv) /bi-ʤaanib/ ثدبت (p.1588)beside

قجش اىج خذيم /xa.diim/(n.) /xaa.dim/ خبد

(p.1604) servant

Group (b) encompasses words of different categories (two nouns and an adverb),

yet they all surface with a shortened vowel. The first /Haaʤ/ is a closed, heavy,

monosyllabic, bimoraic word that is kept bimoraic with VS. The word /Hag/ is

currently popular. Originally, it was used for calling upon someone who did

pilgrimage, then was used for calling elderly people in general. The second word is a

concatenation of a prep. /bi-/ and the adverb of place /gaanib/, which is measured

against the form [FaaʕiL]. The morpheme [ʤaanib], thus, is underlyingly disyllabic,

but surfaces as the monosyllabic, bimoraic [gan(m)b]. What seemed to happen here is

that the high vowel in the nucleus of the ultimate syllable was omitted giving rise to

the form */gaanb/. Nevertheless, the syllable [CVVCC] does not exist amongst the

five syllable types in Arabic listed by Abdil-Tawwaab (2000, p. 55) (see section

2.5.2). Consequently, the long vowel becomes shortened resulting in the form

/gan(m)b/, which is in use at the present time. It could be noted that VS amongst other

phenomena could be a result of people‘s proneness towards ease of articulation, where

this conforms to zipf‘s principle of ‗the least effort‘ or `economy‘, which suggests that

―the ways in which human beings organize their linguistic systems and exert

themselves in speaking tend towards least effort‖ (p. 12).

As for the word /xaadiim/, it is noticed to be one of a kind in the data under

study. The change affects, primarily, the weights of syllables, where /xaa/ vowel

length transferred to the ultimate syllable /dim/, leading to the ultimate syllable‘s


41

alternation to the super heavy /diim/. This occurs at the expense of the penult.

syllable, turning to light /xa/. The form might be a consequence of overcorrection.

The term has been defined by Crystal (2008) as: the phenomenon when ―speakers of

a non-standard dialect attempt to use the standard dialect and ‗go too far‘, producing

a version which does not appear in the standard‖ (p. 232). Thus, in the case viewed

here, speakers might have measured it against such words as adjectives /qadiim/

which eventually gave rise to false ,(long)طو/Tawiil/ ,(weak) ضو/haziil/ ,(old)قذ

analogical form. The VS instances here might be reflecting an idiosyncratic attitude

that lacks regularity, represented in the absence of more similar words throughout the

epic. It could also be regarded as lexical because of the apparent exceptionality,

analogical attitude to other templatic forms, restriction to word boundary and

observing structure preserveration.

Words exhibiting VS have been captured across different consecutive ages,

precisely, during the Fatimid and Ottoman eras. Ahmed (2014) mentioned words

such as ثذسح (p. 123), سحالد ,أقجيل (p. 124), /badrah/ instead of /baadirah/

/and /raHlaat ,(meet you) اقبثيل/ʔaqā(ʔa)blak/ instead of /ʔuqāābilak/ (initiative)ثبدسح

(now riHalaat/), instead of /rāāHilaat/ ساحالد(journeys), which are analogues to group

(a) from the Fatimid era. Aboul-Enein (2013) has, also, extracted such words asػق

/ʕānāq/ instead of /ʕaanāq/ ػبق(hugged) and /Hashaak/ حشبكinstead of /Haashaak/

.(p. 23) (God forbids)حبشبك

3.3. The Conversion of Interdental Sounds

One of the most distinguished characteristics of modern EA is the disappearance

of interdental phonemes and their replacement by dental ones. According to Al-Wer

(2004) ―interdental sounds occur relatively rarely in human languages, and a change

from fricative interdental to a stop is a widely attested phenomenon‖ (p. 21). The

phenomenon could be observed through the orthography, which might mirror how

Egyptians talked at that time.

Table 13

Page no. & Gloss. Original form in

CA

EA word

transcribed Word in

EA

a) /th/~ /t/+ vowel change

fat /thaxiin/ثخ /tixiin/ رخ

(p. 1713) snake /thuʕbaan/ثؼجب /tiʕbaan/ رؼجب

(p.1809) proper

name

/ ʕuthmaan/ػثب ʕitmaan/ ػزب

b) /TH/ ~ /d/

(p.1608)

take

/xuTH/خز /xud/ خذ


42

(For more examples, check appendix 1 no.(d))

Words in paradigm (a) are nouns, which commonly have an interdental /th/.

However, the interdental /th/ is noticed to be replaced by the dental /t/, which is the

closest amongst plosives to the /th/ place of articulation. As for the word /ʕuthmaan/,

it is a proper noun that denotes a male‘s name and the interdental could be postulated

to have turned into the plosive sound (i.e. also pronounced currently as /ʕuSmāān/,

where other than /t/, /th/ also alternates to /s/ or /S/). Paradigms (b) and (c) both

manifest an alternation of the voiced, fricative, interdental /TH/ into a plosive at one

instance, and into a fricative at another. Group (b) has /TH/ surfacing as /d/, where

the phoneme is observed to occur initially, medially, and finally. Thus, the same

counterpart persisted in all instances, which opposes the interdental alternation into

fricative in group (c). The voicing feature is noticed to be preserved in the course of

the conversion of interdentals attested in the data above. In addition, the variation

taking place aims at getting rid of the interdental feature, where the sibilant manner is

sometimes preserved (i.e. /TH/</z/). Alternation of interdentals now takes place with

all Arabic interdentals, with no restrictions to cerain phonological or morphological

governing conditions. Another noticeable feature is its transference to L2 (postlexical

feature), which could be observed with some English learners, for example, when

they tend to alternate the interdental fricatives into /s/ and /z/, e.g. /za/ for ‗the‘, and

/srii/ for ‗three‘. This might provide an evidence for the phenomenon postlexicality.

The phenomenon has been assumed as indigenous for Egyptians, who spoke

Coptic before Arabic (Ahmed, 2014, p. 43). This view was enhanced by the fact that

ancient Egyptian as well as Coptic lacked interdental sounds, as they encompassed

solely one interdental phoneme (i.e. the one known as theta (Θ) in Coptic pronounced

as /th/). However, the interdental phoneme was said to have been replaced by /t/ as

was discovered in some words, particularly in the Middle kingdom sources of ancient

Egypt (Māqāār, 2006, p.194). This view, thus, is generally based on believing in the

indigenous language extended influence (i.e. traces of native tongue). Another

interpretation attributed the influence to the contact between Egyptians and some

Arab tribes, where Al-Tonsi, Al-Sāwi & Massoud (1987) mentioned that the tribe of

melted /THaab(a)/راة /daab/ داة

(p.1517)his chin /THāqnuh/رق /daq(ʔ)nuh/ دق

(p.1722) threw /HaTHaf/حزف

/Hadaf/ حذف

c) /TH/ ~ /z/

(p.2283) excused /maʕTHuur/ؼزس /maʕzuur/ ؼضس

(p.2741) seed /biTHrāh/ثزسح /bizrah/ ثضس


43

Khāyber خجش used to replace /th/ with /t/. In addition, turning the /TH/ into /z/ was a

speech characteristic of Rābii‗a tribe سثؼخ (p. 6).

The phenomenon was detected in Fatimid era, where Ahmed (2014) detected

from Al- Maʤmuu‘Al-Safawiy by ‘Ibn Al-‗Assaal such words as "مزش" ,"رب ‖ (p.

85) /ʔa(i)l- taamin/ instead of /ʔal-thaamin/ اىثب(the eighth), /k(a/i)tiirah (a lot of),

instead of /kathiirah/ مثشح. In addition, the phenomenon accentually showed during

the Ottoman period, where Zack (2009) declared that ―the /th/ was pronounced as /t/

in the dialect of (Ottoman) Cairo‖. She exhibited some examples as a proof from

Daf’ Al-‘ISr such as ―/nadl/‖ ―ذه(wicked) instead of /naTHl/زه, /yiHadrim/حذس (to

speak quickly) instead of /yuHaTHrim/حزس, also /tifl/ (dregs)رفو instead of /thifl/ثفو"

(pp. 89, 90). Furthermore, Aboul-Enein (2012) extracted such words as ―/ladaʕituh/

(burned /cauterized him) for /laTHaʕatuh/, in addition to /ʔillat/ اىيذ(kneading),

instead of /ʔillath/ اىيث (used in the current colloquial expression ʔillat wil ʕagn اىيذ from ‖ثؼثش/for /bāʕthār (scattered)ثؼزش /meaning extra, useless talk), and /bāʕtār ,اىؼد

Al-MuqtaDāb (pp. 35, 36).

3.4. Processes Affecting Consonants: Emphasis Spread

Emphatic sounds االطاد اىفخخ constitute a set of sounds, which particularly

distinguish Semitic languages in general, henceforth, Arabic language and its

dialects. According to Crystal (2008), an emphatic phoneme is ―a type of consonant,

associated particularly with the Semitic languages‖, adding that it ―is articulated in

the pharyngeal or uvular regions of the vocal tract‖ (p. 167). Thus, they form a

natural phonemic class in CA (/T, D, TH, S/) as well as in EA (/T/, /D/, /Z/, /S/). In

addition, some sounds, which are not underlyingly emphatic would become

emphasized by virtue of neighboring an emphatic phoneme or another emphasized

one. Davies (1995) asserts that ―in Arabic dialects, when an underlying emphatic

occurs in a word, emphasis (i.e.pharyngealization) typically spreads to the

neighboring sounds‖ (p. 466). Bin-Muqbil (2006) adds also that the set of uvulars

(i.e. /q/, /x/, /gh/) could spread emphasis to adjacent vowels (p. 45). Another class

includes ―independent secondary emphatics‖ (Harrell, 1956, p. 72). Watson (2002)

has listed out the four EA velarized allophones that belong to this category as: the

laterals /r/ and /l/, and the bilabials /b/ and /m/ (p. 21).

Looking up the following words in bahith.net, it was observed that they do not

exist under the same orthographic representation, where, in opposition to EA words, no

emphasis spread was evident in CA surface forms. The words witnessed throughout the

epic could be said to reflect the conversion of CA words to colloquialism by virtue of

emphasis spread as viewed below:


44

Table 14

The data in group (a) and (b), in general, reflect the influence of emphasis spread

on underlyingly non-emphatic phonemes. The same secondary emphatic sound has

been located with all the data, namely, the independent secondary velarized allophone

[r]. Hence, the velarized, allophonic [r] shows to be the source from which emphasis

transferred to neighboring phonemes. That is to say that the emphasis feature held by

[r] has spread towards the fricative /s/ in (a), and towards the plosive /d/ in (b),

turning them to their emphasized correspondences /S/&/D/. Emphasis spread could be

observed as moving forward, i.e. progressively, in these examples. Moreover, the

process in (a) is observed to take place within the same syllable boundary, with the /r/

occupying onset and the /s/ occupying coda positions. Nevertheless, the trigger [r]

and the affected consonant /d/ are located in neighboring syllables, i.e. /rā.Diy.yāh/,

i.e. heterosyllabic.

The following table manifests an opposite direction of emphasis spread to that

attested in group (13) as is revealed through the following data:

Table 15

Page no. & Gloss. Original form

in CA

EA word


a) Progressive Emphasis Spread ( /s/< /S/)

(p. 1703) (stop talking,

imp.)

/ʔixras/اخشط /ʔixrāS/ اخشص

(p. 1703)

(guards)

/ʔal

Hurrāās/اىحشاط /ʔa(i)l HurrāāS/ اىحشاص

(p. 1726) (to

guard)

/yaHrus/حشط /yuHruS/ حشص دل

b) Progressive Emphasis Spread ( /d/ < /D/)

(p.2283) bad /rādiiʔah/سدئخ /rāDiyyāh/ اىضاخ سض

Page no. &

Gloss. Original form

in CA

EA word



45

(For more examples, check appendix 1 no.(e))

The velarized (i.e. emphatic) [r] in paradigm (c) is evident to affect the fricative

/s/, turning it to its emphasized correspondent /S/. The different direction of both

locations of sounds has not prohibited the process, as here /s/ is situated before the

secondary emphatic [r]. As for inspecting the emphasis spread domain of application

in terms of syllable structure, it was detected that the application took place within, as

well as across syllable boundaries. In addition, the word /tixāDDār/ in (d) is similar to

/rāDiyyāh/ in terms of the affected sound /d/</D/. However, it fits with table (14),

because of the regressive direction of emphasis it manifested. In addition, one

important general characteristic that is noticed as connected to the emphatic /r/ is its

recurrence with the back vowel (i.e. /ā/). Another postulation could be attributing the

source of emphasis spread to the uvular /x/ in both /ʔixrāS/ and /ti(Ti)xāDDār/.

Harrell (1956) mentioned few words that have an emphatic /x/ in EA, such as

/mufāxxām/ and /fuxxāār/ (p. 76). The emphasis scope here, therefore, might be

extended further to affect the spotted pharyngealized consonants in both words.

Emphasis spread in association with the data displayed above is exclusive to

word boundary and only shows to surpass the syllable domain. Harrell (1957) argued

that emphasis spread in EA is a gradient feature, which means that the degree of

velarization is controllable by the speaker. In addition, it was also pointed out as

dependent on style, where he even related it to gender with more tendency of females

to alleviate emphasis (pp. 81- 82). Watson (2002) agrees stating that ―in contrast to

lexical processes, post-lexical processes often result in gradient rather than binary

outputs: in the spread of pharyngealization …, emphasis decreases towards the end of

the domain and may result in a segment which is pharyngealized at the edge close to

the trigger, but non-pharyngealized at the edge away from the trigger‖ (201).

Moreover, Harrell purported that it is dependent on style, where he even related it to

c) Regressive Emphasis Spread ( /s/</S/)

(p. 1610) walked /saar(ā)/عبس /Sāār/ ث اى يل عظ صار

(p. 1502) fence (s) /ʔal-

suur/اىغس

/ʔaswaarāna/األعاس

/ʔa(i)l-

Suur/ /ʔāSwārnā/

أطاسب -اىظس

(p. 1811) palace /sarāāyah/عشاخ /Sārāāyāh/ طشا

Release /sarāāHi/عشاح /SārāāHi/ طشاح

d)Regressive Emphasis Spread (/d/ < /D/)

(p.1592)

(anesthetize) /tuxaddir/ذس رخضش /ti(Ti)xāDDār/ رخ


46

gender with more tendency of females to alleviate emphasis (pp. 81- 82). These

reasons affirm the postlexicality of the process.

This process has been witnessed in EA since the Fatimid era. An evidence of its

antiquity could be witnessed in Ahmed (2014) mentioning words as "شقض",‖سفض‖,

,rāfāS/ (kicked), /murquS/ (proper name), /Suur/ (fence)/ ;(pp. 28, 29) ‖طس― ,"اخشص―

/ʔixrāS/ (be quiet), and /gārāS/ خشص(bell) instead of their CA versions, with /S/

instread of /s/. Aboul-Enein (2012) has extracted such words as ―/Sihriig/

to walk)بض /as well as /māāyiS ,(acts merrily) ض/yihāyyāS/ ,(water tank)طشح

loosely)‖ from Al-MuqtaDāb, also with /S/ instead of /s/ (p. 36). Furthermore, the

phenomenon is widely used at the present time, and is a source of attraction for many

modern linguists.

4. Results and Findings

The field under which the current study was conducted, namely sociophonetics,

has proved appropriate, as it could manage to deal harmoniously with both the

sociohistorical dimension and the linguistic dimension as well. In addition, lexical

phonology was capable of dealing with the morphophonological alternations

encountered via organizing and classifying processes into the suitable components of

grammar, based on the data available whether historical or present day ones.The

moraic model was, also, beneficial, especially in cases of representing weights of

syllables and the changes accompanying length of vowels. Thus, it proved helpful in

representing and accounting for particular phenomena.

The data comprised 140 EA words exhibiting varied altenations, such as glottal

stop deletions, vowel shortening, varaitions of glottal stop, conversion of interdentals,

and emphasis spread. The highest marked variation was generally detected in the stop

alternations with a total number of 104 words, amongst them final stop deletion after

a long vowel recorded the highest frequency of occurrence with 38 words, whereas

final glottal stop deletion after a short vowel had an occurrence of only 14 words. The

second highest alternation was identified in stop alternation into the glide /y/ with a

total number of 24 words. The rest of words manifesting other kinds of alternation

had a total number of 36 words, with the higest number detected through words

reflecting conversion of interdentals (17 words), followed by emphasis spread (11

words), and finally comes vowel shortening with an occurrence of 7 words.

Almost all of the selected sound alternations addressed in this paper were

proved to reflect a regular attitude, which also implied that they are rule-governed.

However, some words reflected an idiosyncratic behavior, which was shown by the


47

absence of a regularity of sound alternation. The systematic behavior was noticed as

a characteristic feature of final glottal stop elisions in a word final position, provided

it is preceded by either a long vowel (table 1) or a short vowel of different qualities

(table 2). The process reflected an affiliation to the lexical module, as it took place

within word boundary, in addition to exhibiting signs of non-finality, i.e. other

changes might take place after suffixation. This designated the probability of other

processes to follow the deletion process in an interactional manner (i.e. cyclicity).

The cyclicity was evident in the case of suffixation to consonant initial morphemes

leading to stem final vowel lengthening, e.g. /ʤazaaʔahu/خضاء</g(j)azaah/خضا (table

3). In addition, /h/ was depicted to be related to final stop deletion after a long

vowel. It reflected sensitivity to the phrasal location, where word final position in the

data triggered /h/-insertion; hence, it proved to be ordered after lexical processes, i.e.

postlexical. In addition it was subsumed to be close to ‗Haaʔ Al-Sakt‘ attested in CA

phonological rule.

Alternation of the glottal stop was located in a word/stem medial position. The

medial stops attested underwent changes, which range between a deletion and an

alternation. The change was determined by the surrounding phonological

environment, where the data revealed that the stop would be deleted, when preceded

by a short vowel and followed by a consonant, e.g. /tastaʔhil/ < /ta(i)staahil/ (table 4).

There are signs of cyclicity, which is reflected in the deletion followed by the vowel

lengthening. It has lexical exceptions that could be viewed in synchronic data, such as

/maʔzuun/ (marriage official) أر . The deletion process did not result in a novel

structure (i.e. templatic forms). Moreover, deletion is witnessed in case the stop was

surrounded by short vowels, e.g. /kaʔannak/ < /kaa(n)nak / (table 5). The rule in the

group of table (5), nevertheless, seems to have been lost, which is revealed in

retaining the stop synchronically. It is restricted to word boundary and is structure

preserving. These reasons might provide evidence of the lexical domain of the rule

application.

The data also reflected the alternation of the stop into a high glide, in case it is

preceded by a low, long vowel, and followed by a high, short, front vowel (table 8

and 9). Hence, the conditional phonological context was also observed to coincide

with certain morphological categories derived from similar root patterns, such as the

case with glottalized active participles, i.e. [FaaʕiL], and glottalized broken plurals

i.e. [FaʕiiLa(h)], [Fiʕaala(h)]. In addition, the alternation into /y/ was revealed also in

the environment C_V, e.g. /mashʔuum/ < /mayshuum/ (table 10).

Most of the changes related to glottal stop have, most probably, occurred for

varied reasons. The resaons could be, first, ease of articulation or resorting to


48

economizing effort, which also agrees with Zipf‘s law, and second, profound

influence by Egyptians contact with Arab tribes either through commerce or due to

their constant abiding in Egypt. Nonetheless, the effect might have originated,

specifically, from the people of Hiʤaz, who were known for traditionally forsaking

the glottal stop realization.

As for initial glottal stop epenthesis, it reflected higher regularity with

imperative, triliteral verbs (table 7) rather than with perfectives (table 6). This is

because, in the case of glottal stop epenthesis in imperatives, the phenomenon seems

to be originating from Hamzatu-l-WaSl tradition in CA. The major difference lies in

Egyptians deriving the imperative verb from a colloquial version of imperf. form. EA

imperatives would have a glottal stop in initial position in analogy to CA rules of

Hamzatu Al-Wasl, e.g. /ʔitsalla/ from /ya(i)tsalla/ in EA. The phenomenon is

conditioned by the verb position within the phrase, i.e. initial, which renders the

process to the postlexical module. However, such sensitivity was not observed in the

case of perfective verbs. Hence, this group seems to be equivalent to CA causative

verbs having the same templatic form [ʔaFʕaL], yet the group of verbs captured in the

epic proved colloquial.

One group of words (table 11) with vowel shortening manifested a regularity

affirmed by the phonological environment. The process operated basically on nouns

measured against the nominal templatic measure [Faaʕila(h)], as it operates on the

long vowels in the super heavy, non-ultimate syllables, [CVVCC] after HVD. It

prevents the occurrence of a forbidden trimoraic syllable. VS reliance on HVD is an

evidence of cyclicity. In addition, the restriction to the word level, Structure

Preserving, and the absence of sensitivity to the phrase level within the data

inspected might be indications of affiliation to the lexical module. However, the

other group of words manifesting VS, though they primarily showed no evidence of

postlexicality, yet if the repeatedily attached [f] and the conditional /ʔiTHa/ were

considered, they might then be considered as postlexical (table 12).

As for the conversion of interdentals (table 13), it seems to apply across the

board, with no exceptions. Moreover, it is transferable to L2, which could be

observed with some English learners, for example, when they tend to alternate the

interdental fricatives into /s/ and /z/, e.g. /za/ for ‗the‘ /srii/ for ‗three‘.This might

provide an evidence for the phenomenon postlexicality. Furthermore, it showed to be

one of the historical sound changes, which possibly originated from either the Coptic

as an indigneous language or a dialectal influence from Arabic tribes. The alternation

has not been witnessed to trigger further processes inside the word domain.

With respect to emphasis spread, the [r] reflected a high frequency of

occurrence with a wide spread assimilatory influence. Moreover, the uvular /x/ was


49

suspected as a trigger of emphasis spread with two words /ʔixrāS/, /tu(i)xāDDār/.

The emphasis feature carrier was witnessed to spread to adjacent consonants in both

positions, namely, when the trigger and the emphasized exist within the same

syllable and across different syllables. The assimilation (emphasis spread) was also

witnessed to move progressively (table 14) and regressively (table 15). Furthermore,

it exhibited sensitivity to style, (i.e. degree of formality), thus revealing signs of

postlexicality.

The social dimension of the study is revealed on two levels. The first is drawn

from the fact of the study being conducted on a folk epic that is recited for Egyptians

and by Egyptians at the Mamluk and Ottoman epochs. This, therefore, suggests that

the vernacular employed throughout the epic reflects everyday-Egyptian Arabic

spoken at that time. The second level of the social angle is drawn from the attempts

of unearthing the sociohistorical triggers of each of the phenomenon that were

addressed throughout the research. Hence, this is evident, particularly, in tracing the

origin of glottal stop alternations, as well as interdentals‘ conversion, languae

contact, in addition to the internal force detected through systemacity of application.

Other phenomena seem to have stronger linguistic internal motivation, such as

attested in vowel shortening, i.e. closed syllable shortening, and emphasis spread, i.e.

assimilation.

In sum, EA words were observed to be originally derived from CA, yet with

exhibiting deviations. Some of the detected phonological and morphological

alternations had origins in CA, yet are applied differently, leading to colloquial

versions of CA words, e.g. Hamzat Al-WaSl & vowel shortening. Nonetheless, the

changes attested did not show drastic changes from the Underlying CA. In addition,

the majority of the selected phonological and morphological alternations examined in

this research have reflected antiquity as well as continuity till present-day Egyptian

Arabic. Further, Al-Zāher Beibers could be considered as one of the most important

sources that both preserved and represented Egyptian Arabic dialect spoken at the

Mamluk and Ottoman periods.

5. Recommendations for Further Research

Generally speaking, more researches should be devoted to tracing the

development of Egyptian Arabic throughout history. This could be accomplished via

inspecting for historical documents that are fully or partially written in the colloquial

variety. This, therefore, would contribute to revealing the shape of EA curve of

linguistic change through time. One kind of studies might be, for example, applying

recent linguistic theories to the historical data.


51

Another suggestion is to start an electronic database, which is concerned with

storing all historical documents written in EA. This would lead to building up an

electronic bank of corpora that would preserve such documents as well as provide the

researchers with an easy access to the data required as well as an easy way of

processing such data. Hence, the availability of the sources of data would encourage

researchers to start applying modern linguistic theories to the data. Hence, they could

investigate corpora by integrating both linguistic and extra-linguistic dimensions,

which would provide a more comprehensive perspective.

References

-Primary Sources

Al-Ghitāāni, G. (ed.). (1996). Mawsuu‘at Al- Zāher Beibers. Cairo: Al-Hay’ah Al-

MāSriyyāh lil- Kitaab.

-Secondary Sources


51

Abdil- Tawwaab, R. (1996). Mushkilat Al- Hamza-l- Arābiyya. Egypt: Maktabat-l-

Khangi.

Abdil- Tawwaab R. (2000). LaHn Al-‘amma wa-l- taTāwwur-l- Lughawiy (2nd

ed.). Cairo:

Egypt: Maktabat Zahrāā‘ Al-Shārq.

Abdil- Tawwab, R. (2007). Allahʤah Al- ‗amiyyah Al-MāSriyyāh fi-l-Qārn Al-

Haadi-'shr Al- Hiʤriy. Maʤalat Magʤma’ Al-Lughati-l-‘Arābiyyah, 28, 390-

405.

Aboul- Enein, H. (2013). A Sociophonetic Study of Selected Words in "Al-Qāwl Al-

MuqtaDāb Fiima Waafāqā Lughat Ahli Misr" as Representative of Egyptian

Arabic in the Early Ottoman Period. Occasional Papers, 56, 1-55.

Abu-Mansour, M. H. (1992). Closed Syllable Shortening and Morphological Levels.

In E. Broselow, M. Eid, J. McCarthy (Eds.), Perspectives on Arabic

Linguistics: Papers from the Annual Symposium on Arabic Linguistics

(Vol. IV), (pp. 47-76). Amesterdam, Philadelphia: John Benjamins

Publishing Company.

Ahmed, A. S. (2014). Al- Lahʤah Al- MāSriyyāh-l- Fāāṭimiyyah: Dirāāsah

Taariixiyya WāSfiyyāh

(2nd

ed.). Cairo, Egypt: Al- Akaademiyya Al-Hadiithah lil-kitaab Al-

ʤaami‗iy.

Aitchison, J. (2001). Language Change: Progress or Decay? UK: Cambridge

University Press.

Algeo, J. (2010). The Origins and Development of the English Language. USA,

Michael Rosenberg.

Al- Māqriiziy, T. A. (2007). ’Ighaathat –l- ’Umma bi- Kashf Al- Ghumma. Kārām H.

f. (ed.). Cairo: Ein for Human and Social Studies

Al-Pop, M. I. (2010). Al- Zāāhira Al- SawTiyya ‗ind Sibawayh. Maʤalat Dirāsaat fil

Lughah-l- ‘Arabiyya Wa- Adaabiha, 2, 18- 28.

Allen, J. P. (2004). Middle Egyptian: An Introduction to the Language and Culture of

Hieroglyphs (7th

ed.). UK: Cambridge University Press.

Al-Tonsi, A., Al-Sāāwi, L., & Massoud, S. (1987). An Intensive Course in Egyptian

Colloquial Arabic. Cairo: American University in Cairo.

Al-Shirbiniy, M, K. (1890). Hazz Al-QuHuuf bi-QaSiid Abu-Shaduuf (2nd

Ed.).

Cairo, Egypt: Al- MaTbA‗ah Al-Amiiriyah.


52

Al- Wer, E. (2004). Variability Reproduced: A variationist view of the [D]/ [D]

Opposition in Modern Arabic Dialects. In M. Haak, K. Versteegh & R. Dejong

(eds.). Approaches to Arabic Dialects: Collection of Articles presented to

Manfred Woidich on the Occasion of his Sixtieth Birthday, Amsterdam: Brill,

pp. 21- 32.

‗ĀTiyyāh, A. (2001). Al-Siira Al-Sha‗biyya lil- Zāher Beibers: hal yumkinuhaa ‗an

takuun māwduu‘ li- dirāāsah ‗ilmiyyah. Al- Ārābi. 507 (12), (n.p.). Retrieved

from http://www.alarabimag.com/Article.asp?Art=4040&ID=165

Attinasi, J. J. (1974). The Sociolinguistics of William Labov [Review of Bilingual

Review]. Jstore, 1(3), 279-304.

Badawi, E., Carter, M. G., & Gully, A. (2004). Modern Written Arabic: A

comprehensive grammar. London and New York: Routledge.

Borowsky, T. (1989). Structure Preservation and the Syllable Coda in English.

Natural Language and Linguistic Theory, 7, 145-166.

Bassiouney, R. (2009). Arabic Sociolinguistics. Great Britain: Edinburgh University

Press.

Berendsen, E. (1986). The Phonology of Cliticization. Dordecht, the Netherlands:

Foris Publication.

Bin-Muqbil, M. S. (2006). Phonetic and Phonological Aspects of Arabic Emphatics

and Gutturals. (Unpublished doctoral dissertation). University of

Wisconsin, Madison.

Borowsky. T. J. (1989). Topics in the Lexical Phonology of English. Natural

Language and Linguistic Theory, 7, 145-166.

Broselow, E. I. (1976). The Phonology of Egyptian Arabic (Unpublished doctoral

dissertation). The graduate school of the University of Masachussetts,

Masachussetts.

Carr, P. (1993). Modern Linguistics: Phonology. Great Britain: The Macmillan

Press.

Crystal, D. (2008). Linguistics dictionary (6th

Ed.). UK: Blackwell Publishing.

http://www.alarabimag.com/Article.asp?Art=4040&ID=165


53

Coulmas, F. (2013). Sociolinguitics: The Study of Speakers Choice. United

Kingdom: Cambridge University Press.

Danesi, M. (2004). A Basic Course in Anthropological Linguistics. Toronto, Canada:

Canadian Scholars‘ press Inc.

Davis, S. (1995). Emphasis Spread in Arabic and Grounded Phonology. Linguistic

Inquiry, 26(3), 465-498. Retrieved from

http://www.jstor.org/stable/4178907

De Lacy, O. D. (1923). Comparative Grammar of the Semitic languages. London,

UK: Steiihen Austin d Sons, Ltd.

Dos, M. & Davies, H. (2013). Al- Aamiyyia Al-MaSriyya Al-Maktuubah. Cairo: Al

hay‘ah Al- MaSriyya Al- ‗Amah lil kitaabb.

Eisele, J. C. (1987). Arabic Dialectology: A Review of Recent Literature. Arabic

Dialectology: A Review of Recent Literature Al-'Arabiyya, (20), 199-269. Retrieved

from http://www.jstor.org/stable/43191695

Ezzat, Ali (1986). Aspects of Language Study. Cairo: Sphinx Series.

Foulkes, P. & Docherty, G. (2006). The Social Life of Phonetics and Phonology.

Journal of phonetics, 34, 409-438. Retrieved from

http://www.cs.indiana.edu/~port/teach/sem08/Foulkes.Docherty.socl.life.ph

ontc.phonolJPhon2006.pdf

Foulkes, P., Scobbie, J. M., & Watt, D. (2010). Sociophonetics. In W. J. Hardcastle,

J. Laver, & F. E. Gibbon (Eds.). The Handbook of Phonetic Science (pp.

703-745). UK: Wiley Blackwell

Fullwood, M. A. & O‘Donnell T. J. (2013). Learning non-concatenative

morphology. Proceedings of the Workshop on Cognitive Modeling and

Computational Linguistics, 21–27. Retrieved from

http://www.aclweb.org/anthology/W/W13/W13-2603.pdf.

Gadalla, H. A. H. (2000). Comparative Morphology of Standard and Egyptian

Arabic. München, Germany: Lincom Europa.

Guy, G. R. (1996). Preface. In R. G. Gregory, Feagin, C. Schiffin, & D. Baugh, J.

(Eds.), Towards a Social Science of Language: Papers in honor of William

Labov (pp. ix, xiv). US: John Benjamins Publishing Company



http://www.aclweb.org/anthology/W/W13/W13-2603.pdf


54

Haeri, N. (2000). Arabic Sociolinguistics and beyond. Annual Review of

Anthropology, 29, 61-87. Retrieved from


Haikal, F. (2003). Introduction. In Ahmed Y. (Ed.), From Pharaohs Lips: Ancient

Egyptian Language in the Arabic of Today Cairo (pp. vii-xii). Cairo,

Egypt: The American University in Cairo Press.

Harrell, R. S. (1957). The Phonology of Colloquial Egyptian Arabic. New York:

American Council of Learned Societies.

Haywood J.A. &Nahmad H. M. A. (1965). New Arabic grammar. London: Lund

Humphries.

Hilaal, A. H. Allahaʤaat Al- ‘Arabiyya nash’atan wa-taTawuran (1993). Egypt:

Makatabat Wahbah.

Hinds, M & Badawi, E. (1986). A Dictionary of Egyptian Arabic. Egypt: Liberary of

Lebanon.

Ibn Ginni O. (1993). Sirr Sinaa’at Al-I’raab (1st vol.). Hasan A. (ed.). Daar Al-qlam.

Dimascus, Beirut.

Kiparsky, P. (1985). Some Consequences of Lexical Phonology. Phonology

Yearbook, 2, pp. 85-138.

Kiparsky, P. (2003). Syllables and Moras in Arabic. In C. Frey & R. Van de Vijver

(Eds.), The Syllable in Optimality Theory (pp. 147-182). UK: Cambridge

University Press.

Kramer, R. (n.d.). Egyptian. Retrieved from

http://faculty.georgetown.edu/rtk8/Egyptian%20Chapter.pdf

Lane-Poole, S. (1901). A History of Egypt in the Middle Ages (Vol. 6).USA, New

York, NY: Charles Scribners‘ son.

MaSri, M. N. (2009). Al-Istabraq fi Riwaayat Warsh ‘an Naafi’ min Tarriq Al-Azraq

(3rd

Ed.). Retrieved from http://library.tafsir.net/book/4672

http://faculty.georgetown.edu/rtk8/Egyptian%20Chapter.pdf


55

Māqāār, S. (2007). Al- Lahga Al-‘amiyyah wa- Guzuurha- l- MāSriyyāh. Cairo:

Maktabat Madbuuli.

Maye, J. (2001). Trisyllabic Shortening and Two Affix Classes. Coyote Papers, 10,

78-96. Retrieved from

http://arizona.openrepository.com/arizona/handle/10150/311825

McCarthy, J. J. & Prince, A. (1990). Closed Syllable Shortening and Morphological

Levels. In J. J. McCarthy & M. Eid (Eds.), Perspectives on Arabic Linguistics:

Papers from the Second Annual Symposium on Arabic Linguistics (Vol II), (pp. 1-

26). Amesterdam, Philadelphia: John Benjamins Publishing Company.

McCarthy, John J. (1992). Closed Syllable Shortening and Morphological Levels. In

E. Broselow & M. Eid, J. McCarthy (Eds.), Perspectives on Arabic

Linguistics: Papers from the Annual Symposium on Arabic Linguistics

(Vol. IV), (pp. 1- 26). Amesterdam, Philadelphia: John Benjamins

Publishing Company.

McCarthy, J. J. (1999). Sympathy and phonological opacity. Phonology, 96, 331-

399. Retreived from

http://scholarworks.umass.edu/linguist_faculty_pubs/96

McCarthy, J. J. (2007). Derivations and Levels of Representation. In P. de Lacy (ed.),

The Cambridge Handbook of Phonology (pp. 99- 118). NY, USA:

Cambridge University Press.

McMahon, A. M. (1994). Understanding language change. Cambridge: Cambridge

University Press.

McMahon, A. M. (2000). Lexical phonology and the History of English. Cambridge:

Cambridge University Press.

Meyer, C. F. (2009). Introducing English Linguistics. UK: Cambridge University

Press.

Miller, C. & Caubet, D. (2010). Arabic sociolinguistics in the Middle East and North

Africa (MENA). In M. J. Ball (Ed.), The Routledge Handbook of

http://arizona.openrepository.com/arizona/handle/10150/311825

http://scholarworks.umass.edu/linguist_faculty_pubs/96


56

Sociolinguistics around the World (pp. 238-256). New York, NY:

Routledge.

Mitchell, T. F. (1956). An Introduction to Egyptian Colloquial Arabic. Oxford:

Oxford University Press.

Mohanan, K. P. (1986). The Theory of Lexical Phonology. Dordrecht: Reidel

Publishing Company.

Na‗iim, R. (ed.) (n.d.). Al-baheth Al Arabiy. Retrieved from

http://www.baheth.info/.

Newman, D. (2006). The Phonetics of Arabic. Arabic Phonetics: Sound Descriptions.

1-6. Retreived from http://www.dur.ac.uk/daniel.newman/phon5.pdf.

Noske, R. (1993). A theory of Syllabification and Segmental Alternation. Tubingen:

Niemeyer

Ohala, J. J. (1990). There is no interface between Phonetics and Phonology: A

personal view, Journal of Phonetics, (18), 153-171.

Oostendrop, V. (2005). Mora Theory [PDF format]. Retrieved from

www.vanoostendorp.nl/pdf/051122.pdf

Pandey, K. P. (1997). Optionality, lexicality and sound change. Journal of

Linguistics, 33 (1), 191-130 Retreived from

http://www.jstor.org/stable/4176396 .

Ryding, K. C. (2005). A Reference Grammar of Modern Standard Arabic. New

York, NY: Cambridge University Press.

Salah El-Din, M. (2010, August 20). QaamuuS Al- Amthaal Al-Sha‗biyyah Al-

MaSriyya. [Web log comment]. Retrieved from

http://mohdkenzo.blogspot.com.eg/2010/08/blog-post_3369.html.

Selim, M. R. (1957). Al-’Adab Al-‘Ārābiy wa- Taariixuh fi‘ASr-l- Mamaaliik wa-l-

‘Uthmaanyiin wal ‘ASr-l- Hadiith. Egypt: Dāār-l- kitaab –l- Arabi.

Shalaby, M. (1992). Hayaat Al-Malik Al-Zāāher Beibers: Al-Asad-l-Dāāri qāāhir Al-

tataar wa-Mudammir Al-Sālibyiin. Lebnaan, Beirut: Daar Al-Gabal.

http://www.baheth.info/

http://www.dur.ac.uk/daniel.newman/phon5.pdf

http://www.vanoostendorp.nl/pdf/051122.pdf

http://mohdkenzo.blogspot.com.eg/2010/08/blog-post_3369.html


57

Sibawayh, A., I. O. (1982). Al-Kitaab. A. M. Hāāruun (Ed.). Egypt: Maktabit-l-Khangi.

Smrž, O., Bielický, V. R, Buckwalter, T. (2009). ElixirFM Functional Arabic

Morphology. Retrieved from http://quest.ms.mff.cuni.cz/cgi

bin/elixir/index.fcgi?mode=inflect&clip=(417,1).

Stabler, E. (2007). Linguistics: An introduction to the methods and some basic ideas of

theoretical linguistics

[PDF document]. Retrieved from

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.150.749&rep=rep1&type

=pdf

Watson, J. C. E. (2002). The Phonology and Morphology of Arabic. Great Britain:

Oxford University Press.

Wodak, R., Johnstone, B. & Kerswill, P. (2011). Introduction. In R. Wodak, B.

Johnstone & P. Kerswill (Eds.), The Sage Handbook of Sociolinguistics (pp.1-8).

UK: Sage Publications.

Zack, E. W. (2009). Egyptian Arabic in the seventeenth century: A study and edition of

Yūsuf al Maġribī’s Dafʿ Al-iṣr ʿan kalām ahl Miṣr (doctoral Dissertation).

University of Amsterdam, Holland. Retrieved from Academia.com.

Appendix 1

Additional words from Beibers reflecting the phenomena under study

http://quest.ms.mff.cuni.cz/cgi


58

a)

Final

glottal

stop

deletion

after a

long

vowel

(38 words, only

11 with āā)

/kimyaaʔ/مبء</kimya/مب(chemistry)(p.1702),/miinaaʔ/بء</miinah/(harb

or) (p.1488), /Hayaaʔ/ حبء</Haya/ حب(shyness) (p.1555),

/ʔiHyaaʔ/احبء</ʔiHya/ احب (to revive) (p. 1659), /xalaaʔ/خالء</xala/ خال (empty

space), (p. 1509), /duʕaaʔ/دػبء</duʕa/ دػب(prayer),/ghāTāāʔ/غطبء</ghāTā/

,(p.1532)(cover)غطب

/baqa/ثقب</bāqāāʔ/ثقبء(stay)(p.1478),/ʔalʔumarāāʔ/االشاء</ʔalʔumārā/االشا(pri

nces)(p.1640),/bayDāāʔ/ثضبء</beeDā/

ʔal/ ,(p. 302) (black) عدا/sooda/>عداء.sawdaaʔ/,(p.1513)(white)ثضب

wuzarāāʔ/اىصساء</ʔa(i)l wuzārā/ اىصسا(ministers) (p. 1534), / ʔal

radiiʔ/اىشدئ</ ʔa(i)l radi/ اىشد(bad) (p. 2849), /ʔibtidaaʔ/اثزذاء</ ʔibtida/اثزذا (beginning), / ʔintihaaʔ/ازبء</ ʔintiha/ ازب (end), / ʔal ʕishaaʔ/اىؼشبء</ ʔa(i)l

ʕisha/اىؼشب (a prayer) (p. 1506), /huduuʔ/ذء</hudu/ ذ (serenity) ,

/hawaaʔ/اء</hawa/ ا(air) (p. 1593), /ʔal ʔashyaaʔ/االشبء </ʔa(i)l ʔashya/

,(p. 1489) (loyality)اىفب /ʔal wafa/>اىفبء/ʔal wafaaʔ/ ,(p. 1537) (things)االشب

/ʔal nisaaʔ/اىغبء</ʔal nisa/ اىغب (night) (p. 2849),

/ʔaHibbaaʔuh/احجبئخ</ʔaHibbaah/احجب(his beloved) (p. 204),

b)

Final

Glottal

Stop

deletion

after a

short vowel

(14 words)

/yataʤarrāʔ/زدبسأ</yitgāārā/ زدبس(dare)(p. 1641), /yubarriʔ/جشئ</yubri/

,(p.300)جبس

/ʔal baadiʔ/اىجبدئ < /ʔal baadi/اىجبد (the one who begins) (p. 2846). /ʕabbaʔ

/ʕabba/>ػجأ/ ػجب to load(p.1610) /yuqriʔ/قشئ</yuqār(r)ii/or/yu(i)qarri/) قششصػ (make someone read), (p.1577), /ʔinkafaʔ/انفأ< /ʔinkafa/ انفب (to

recede)(p.2220)

c)

glottal stop

alternation

into the

glide /y/

(24 words)

/saaʔiq/عبئق</saayiq/ عبق (driver) (p. 1617), /xaaʔif/خبئف</xaayif/ خبف(

scared)

(p. 1582), /qāāʔil/قبئو</qaayil/ قبو(the one who said), /naaʔim/بئ</naayim/ب (sleeping) (p.1501), /saaʔir/عبئش</saayir/ عبش(the one who walks) (p.1604),

/daaʔir/دائش</daayir/ داش(p.1586) , /ghaaʔir/ غبئش </ghaayir/غبش(hollow)

(p.1616), /rāāʔiq/سائق< /raa.yiq/ʔ/ ساق (someone with merry mode),

/rāā.ʔiH/سائح</raayiH/ ساح (going)(p.1783).

d)

Interdentals

conversion

(16 words)

/ʔawthāqti/ أثقز < /ʔāwsāqtii/ اعقز (constrained) (p. 1647),

/ʤuTHuur/خزس</guzuur/ خضس(roots) (p. 1796), /THurāh/رسح</durāh/ دسح(corn)

(p. 1801), /kaTHa/مزا</keda/ مذا(like this)(p. 1712)

e)

Emphasis

Spread

(11 words)

/SuuT/ طط from /sawT/ عط (whip), /yubaasiT/جبعظ< /yubāāSiT جبطظ (p.

2756)(speak openly).


59

Appendix 2


61

a sociophonetic study of selected morphophonetic ...€¦ · egyptian arabic, from about the mid of...

Documents